AfroTech logo and headshots of  Dela Agbemabiese, Justin Barry, Nashlie Sephus and Colby Wise.
With AfroTech World occurring this week, we asked some of the company's Black scientists what they consider some of the systemic issues limiting underrepresented minorities from being more involved in the technology industry. We heard from Dela Agbemabiese (lower left), a data scientist, Justin Barry (upper left), applied scientist, Nashlie Sephus (upper right), applied science manager, and Colby Wise (lower right), senior deep learning scientist.
Credit: Glynis Condon

Issues of racial, ethnic and gender diversity are on the agenda at AfroTech World

Amazon scientists provide insights on issues related to lack of involvement of underrepresented minorities in the technology industry.

As CNBC reported earlier this year, six years after initially disclosing diversity reports, major technology companies have made little progress in hiring more minorities, especially Black employees with science and technology skills.

This presents a series of ongoing challenges. According the US Bureau of Labor Statistics (BLS), nearly one-quarter of the country’s total economic output is produced by high-tech industries, and in 2017 BLS projected there would be more than 1 million job openings in computer and information technology over the next 10 years. Moreover, computing occupation salaries are more than twice the median wage for all other occupations, according to BLS.

“When we look at tech and its impact on our economy, and the simultaneous underrepresentation of the Black community, it is a critically important racial and economic justice issue," says Allison Scott, CEO of the Kapor Center. “When the tech workforce and leadership reflects the diverse experiences and backgrounds of our nation, I believe tech can begin to play an integral role in addressing long-standing disparities that exist in this country.”

As of December 31, 2019, Amazon reported that 26.5% of its global workforce identifies as Black/African American, 26.5% Asian, 18.5% Hispanic/Latinx, 1.3% as Native American, and 3.6% as two or more races.  The 26.5% of employees who identify as Black/African American work in both non-technical and technical roles.

This week at AfroTech World, issues related to the lack of adequate racial, ethnic, and gender diversity within the technology industry are on the agenda as leaders in technology and business come together to exchange ideas for creating greater opportunity for Blacks in technology.  Amazon is a Diamond Sponsor of this year’s event, and has a virtual recruiting booth.  

On Nov. 13, the company is hosting a virtual event, “Our Voices, Our Power”, presented by Amazon’s Black Employee Network (BEN) affinity group. Attendees will hear employees share their Amazon journey stories, learn about career opportunities, and enjoy entertainment.

In advance of AfroTech, Amazon Science asked some of the company’s Black scientists what they consider some of the systemic issues limiting underrepresented minorities’ involvement in the technology industry, about some of the issues they have had to overcome in pursuing their science careers, who or what inspired them to pursue their science careers, and what lessons we might take from their individual experiences. 

Dela Agbemabiese is a data scientist within Amazon’s advertising organization. He earned his master’s degree in business administration from Drexel University.

Dela Agbemabiese
Dela Agbemabiese

What do you consider some of the systemic issues limiting underrepresented minorities from greater employment opportunities in the technology industry?

Lack of financial resources to stimulate curiosity in tech, lack of mentors or heroes to look up to due to low representation, and societal prejudice hindering opportunities.

Lack of financial resources to stimulate curiosity in tech. I have been fortunate and blessed my entire life.  All gratitude goes to my parents. I was born in Ghana, West Africa. My mom was a nurse, and my dad an economist. Due to the nature of my dad’s work, I got the opportunity to travel a lot as a kid, got enrolled into a course at eight years old to get a Linux command line certificate, and always had access to tech resources. My parents sacrificed to ensure I attended the best schools, and there is not a single thing I ever asked for that I did not get. This may not be the case for all children, whose parents are possibly working hard doing multiple jobs, and in some cases are single parents. If the financial resources I had were similar to that of many minority children, it would be unlikely for me to be where I am today.

"Students don’t see themselves represented in the [economics] profession"

Four economists from diverse backgrounds explain why diversity is essential—and what needs to happen to achieve it.

Lack of mentors or heroes to look up to due to low representation. While my dad was heavy on econometrics and I learned a thing or two from him, it was my cousin Martey to whom I looked up. He was brilliant academically, and I always wanted to be like him. He tutored me in math and physics, thus giving me an edge over my classmates. Martey was not my only mentor, in fact, I had many, including Yao Obeng, who helped me nurture my creativity and problem-solving skills. Many minority children may not have mentors or heroes within tech to encourage and inspire interest in tech-related careers. If I did not have these mentors to motivate me, it would be unlikely for me to be where I am today.

Societal prejudice hindering opportunities. Growing up in Ghana, prejudice did not exist from a racial standpoint. Once I moved to the United States for my undergraduate degree, this became a reality. My minority friends and I have had to work twice as hard as our peers to prove we are as good as our credentials. We strived to invalidate stereotypes about minorities through the quality of our work and our work ethic. With everything I do, in the back of my mind I am thinking about how my actions or inactions affect the perception towards minorities: am I enabling some of these unfounded prejudices? Or am I, through my work, educating my peers and superiors? For me, this societal prejudice only began when I came to the United States for my undergraduate degree, but imagine the minority children out there who have had to live with this their entire lives. It sure can get demoralizing.

What are some of the obstacles you had to overcome in pursuing your science career?

Societal prejudice hindering opportunities. I have been lucky to have managers and peers that are inclusive and open-minded, that judge me based on the quality of my work. Rachel McKitrick was my first manager in Amazon. I joined Amazon as a business analyst, despite my previous role as a senior data scientist. I just wanted to join Amazon! Rachel knew my business analyst role was not ideal, and gave me projects that were science oriented, which ultimately enabled my transition to scientist. My second manager, Monica Wu, always made herself available to chat and made me feel like my voice and opinion mattered. My current team managed by Dauwe Vercamer and Andrew Petschek welcomed me with open arms, gave me opportunities to shine and lead within the team. They provide direct feedback that has made me a much better scientist today.

I have had the privilege of learning from a lot of people. Societal prejudice may be harder to solve for, but I believe a good place to start will be to find means for minority youth to gain access to some of the brilliant minds within the technology industry, be it through some virtual teaching programs, or through some mentoring programs. The prejudice may exist, the financial resources may be sparse or non-existent, but with heroes and mentors to look up to, a child’s imagination can be sparked for what could be.

Who or what inspired you to pursue your science career, and what lessons can we take from your experience?

My dad due to his econometrics background, and my childhood mentors who encouraged me to put math and science ahead of basketball and soccer. Since then, I have had lots of mentors along the way, especially here at Amazon. Individuals such as Leo Razoumov, Pranjal Mallick, Amy Ruschak, John Lafayette, and Oded Netzer, who have helped shape me into a better scientist.

My advice to Black students interested in a STEM career, or other Black scientists is to find mentors, and get them involved in your work. Meet with them once a week for even 10 minutes, and let them influence your work.

Justin Barry is an applied scientist with Amazon’s Prime Video organization. He earned his master’s degree in computer science from the University of Central Florida.

Justin Barry
Justin Barry

What do you consider some of the systemic issues limiting underrepresented minorities from greater employment opportunities in the technology industry?

This is a massive topic with a myriad of associated socioeconomic issues. One issue that jumps to the forefront for me is the schools where leading companies within the tech industry recruit from. Traditionally, these companies have limited their recruitment to top universities where Blacks and other underrepresented minorities comprise a small percentage of the student population. This is beginning to change, but I believe technology companies need to more aggressively expand their recruitment efforts, especially among historically Black colleges and universities (HBCUs).

What are some of the obstacles you had to overcome in pursuing your science career?

One issue is imposter syndrome — the idea that you're not good enough and you’re only in your position because you’ve been given special treatment. Although imposter syndrome is something everyone experiences, I think it’s particularly acute for Blacks given the clear underrepresentation within the technology industry. Imposter syndrome can touch all aspects of your job if you’re unaware, or if you don’t have the tools to deal with it. Not everyone has the tools to deal with it, and I suspect not everyone has correctly identified the problem.

Who or what inspired you to pursue your science career, and what lessons can we take from your experience?

Video games sparked my interest in computer science, and more specifically artificial intelligence. My undergraduate degree is in computer science and math, and machine learning and AI provide the opportunity to apply my computer science and math skills to real-world applications.

Nashlie Sephus is an applied science manager within Amazon Web Services Ai. She earned her PhD in electrical and computer engineering from Georgia Tech.

Nashlie Sephus
Nashlie Sephus

What do you consider some of the systemic issues limiting underrepresented minorities from greater employment opportunities in the technology industry?

Imposter syndrome is one issue I find common within underrepresented minority groups. It’s a feeling of being convinced that you don’t belong in the industry, or within advanced roles in the industry, regardless of your accolades and accomplishments. It is as if they are not real or didn’t happen. This is often due to not seeing many others who look like you in similar or higher positions. ‘You can’t be what you can’t see’ is a common thought. Also, there are few mentors or support systems for these groups, and as a black/female/engineer/scientist, you sometimes feel like the minority of the minority, which further isolates you.  

What are some of the obstacles you had to overcome in pursuing your science career?

At times, I have had to fight for myself and members of my teams for equal pay and advancement in my career. I also have needed to develop mechanisms to be heard when it was difficult to convey messages to those around me. I’m usually quiet and reserved, but over the years I’ve learned how to gain respect from peers by being more outspoken even, or especially, when I disagreed. This is one reason why I appreciate Amazon’s leadership principle: Have Backbone; Disagree and Commit. 

Who or what inspired you to pursue your science career, and what lessons can we take from your experience?

I grew up in a house full of women where we often did our own chores, like fixing and repairing things around the house. I was also always going to summer math and science camps in elementary and middle school, especially a summer engineering camp for girls after my eighth grade science teacher recommended I attend. This was when I was first introduced to the various areas of engineering, and fell in love with computer science. Being able to control the hardware with software was fascinating to me. I knew then that’s what I wanted to do. This early exposure to science was key to me figuring out one of my passions, in addition to music and sports.

Colby Wise is a senior deep learning scientist and manager within the AWS Machine Learning Solutions Lab. He earned his master’s degree in computer science from the Columbia University Fu Foundation School of Engineering and Applied Sciences.

Colby Wise
Colby Wise

What do you consider some of the systemic issues limiting underrepresented minorities from greater employment opportunities in the technology industry?

Educational opportunity. Science, technology, engineering, and math (STEM) careers in the technology industry are highly competitive. Over the years, we’ve seen advanced tools and technologies like cloud technology, machine learning, and deep learning, that were once reserved only for large companies or prestigious universities being utilized by students as early as junior high school. While this has created and accelerated educational opportunities for millions of students globally, the reality is that not all have been able to benefit. In the United States, public school funding varies significantly by geography, and where you grow up is a major factor in access to educational resources. Schools with advanced STEM courses and other after-school programs are valuable inroads for STEM students to accelerate their learning opportunities and explore careers in science. What’s more, these opportunities compound positively from lower educational levels to higher educational levels. While not the only factor, these programs are important when understanding the pipeline of underrepresented minorities in highly competitive industries like technology. For example, the US Federal Reserve conducted a study highlighting how educational attainment of parents plays an important role in children’s educational pursuits. Studies like this and others indicate that lower parental educational attainment may present a unique challenge for students. One potential consequence of underrepresentation of minorities in advanced degrees is that employment opportunities often arise from one’s social network, employee referrals, for example. This can be summarized as both an employment funnel problem and a network problem. While not always the case, a more diverse workforce can build connections to underrepresented talent pools. 

Financial equality. In a study from 2020, the US Federal Reserve found large and persistent gaps in net wealth and earnings by race and ethnicity. While education is a significant factor in wage gaps, the St. Louis Federal Reserve found net wealth by race was not as positively correlated with educational attainment for minorities. Educational attainment is extremely important. Many highly technical roles require advanced degrees. Financial equality and opportunity as characterized by job salary prospects, current income and net wealth, and access to educational funding sources like loans are all potential factors impacting lower minority employment. In 2016, the Brookings Institution found the median household net wealth for Black and Hispanic families to be 1/8th  that of white households. When you consider the rising cost of college and advanced degrees, this income and net wealth gap may also play a factor in why employment among underrepresented minorities is lower in highly competitive industries like technology. Specifically, minorities whose households cannot readily pay for advanced degrees choose between the implications of high debt burdens and lower comparative earnings, and often must forsake advanced degrees to enter or stay in the workforce.

Leadership representation. Representation of minorities in leadership positions is relatively low. It is unclear how much educational opportunity and financial equality contribute to this, compared to other issues such as equitable pathways to senior leadership positions. In many companies in which I have worked, you notice a similar triangular pattern of minority leadership where representation at junior levels is more in-line with industry trends, while there is a dearth of representation as you reach more senior positions. No doubt there is work to be done to drive greater employment of underrepresented minorities at all levels. But simply increasing the representation at entry levels does not address other attrition and talent-retention hurdles. Overall, companies need to take a more systematic, data-driven approach to move the needle and find solutions to underrepresentation of minorities in the tech industry. For instance, companies should not be afraid to tackle the complex issues at multiple hierarchies, such as creating innovative solutions to drive educational opportunity while objectively measuring current pathways to employment within the tech industry. Furthermore, companies should ensure financial equality by aligning corporate incentives with fair pay distributions, minority leadership representation, and talent development and retention.  

What are some of the obstacles you had to overcome in pursuing your science career?

Educational opportunity. While everyone’s path is different, unfortunately my story is rather common given its similarity to those of many underrepresented minorities. I faced and overcame obstacles in educational and financial opportunity plus roadblocks to leadership roles. I attribute my luck mainly to the many individuals who provided a helping hand, plus a little bit of hard work sprinkled in. I grew up in a single-parent household in an impoverished, high-crime inner-city area. Despite this, my family valued education highly, and one of my parents had an advanced degree which was extremely rare for the area. Given that, I always ranked in the top 1% in my coursework while very young. That said, district educational attainment rates were low, and advanced coursework or programs for gifted students were nonexistent. However, prior to high school an unfortunate family event led to me moving from one of the poorest areas in the country to one of the best school districts nationally. After discovering how far behind I was in math and science, my family and I worked extremely hard over several years to get me back in line with my expected academic grade level. Now fast-forwarding to college: I, like many other minorities, did not have the means to pay for college, nor easy access to loans. After being selected to a number of great schools, my decision was ultimately driven by the amount of money I received in scholarships and grants. During college I followed the same recipe for success: tons of luck, humility to ask for help, and a bit of hard work to land an internship as a sophomore at a prestigious Wall Street investment bank. There I was surrounded by some of the smartest minds in STEM, with many having achieved advanced degrees from top universities around the world. The vast majority of these individuals did not look like me. Desperately wanting to be accepted and succeed among my peers in industry is what drove me to pursue a career in science, and many years later brought me to AWS.

Who or what inspired you to pursue your science career, and what lessons can we take from your experience?

Family and friends. Ultimately, doing what you love and constantly learning while being curious is the greatest inspiration one needs to pursue a career in science. As discussed above, studies have shown a correlation between parental educational attainment and children’s attainment. Thinking forward a bit, I combined my passion for what I love in science — AI/ML — with a selfish goal of wanting to be a living model for a career in science for my children. My greatest inspiration, however, is my wife. She discovered her passion for science at a very young age with plentiful opportunities to explore that passion, ultimately helping her reach the pinnacles of academia, where she received undergraduate and graduate degrees from two of the top universities in the world. Her passion for science, hard work, and humility continue to inspire me on a daily basis.

Related content

US, WA, Seattle
As part of the AWS Applied AI Solutions organization, we have a vision to provide end user applications, leveraging Amazon's unique experience and expertise, that are used by millions of companies worldwide to manage day-to-day operations. We will accomplish this by accelerating our customers' businesses through delivery of intuitive and differentiated technology solutions that solve enduring business challenges. We blend vision with curiosity and Amazon's real-world experience to build opinionated, turnkey solutions. Where customers prefer to buy over build, we become their trusted partner with solutions that are easy to adopt and easy to use. The Team Join the next science revolution at AWS Life Sciences Applied AI Solutions, where you'll work alongside world-class scientists to build AI that transforms how therapeutics are discovered, developed, and brought to patients. We're out to revolutionize how medicines are discovered, developed, and brought to patients, powered by a new generation of AI. Our team tackles some of the hardest open problems at the intersection of frontier AI and life sciences. We apply biological foundation models, large language models, and agentic reasoning systems to life sciences problems, then put them into the hands of pharma, biotech, and diagnostics customers as applications and managed services they can fine-tune, tailor, and deploy on their own data. The science challenges are deep: how do you design agentic systems that reason correctly over complex biological, regulatory, and clinical logic? How do you enable customers to tailor foundation models to their proprietary data and get better outputs with less effort? How do you adapt models to reason faithfully in high-stakes scientific and regulatory domains? Today we're focused on two areas. In drug design, our products (including Amazon Bio Discovery) accelerate discovery by giving bench scientists AI-guided protein engineering and antibody design capabilities. In clinical trials, we're building AI that automates and optimizes regulatory and clinical development workflows. We combine frontier research with production-scale delivery to put breakthrough science into the hands of customers solving humanity's hardest problems. We value scientific rigor, encourage publication, and support conference participation. If you want to do research that ships, this is the team. The Role We are seeking an exceptional Principal Applied Scientist to set the scientific direction for our life sciences AI portfolio. You will be the scientific leader who defines research agendas, architects novel approaches, and delivers models and methods that give our customers capabilities that did not previously exist. This is a rare role that combines deep expertise in LLM reasoning and agentic AI with applied impact in life sciences. You will innovate on how large language models reason, plan, and act in complex scientific domains, while applying domain knowledge in biology to ensure models produce scientifically valid outputs. The problems span multiple fronts: - How do you build LLM-based agentic systems that correctly reason over clinical protocols, regulatory standards, and complex multi-step scientific workflows? - How do you develop model customization and training methods that let customers get state-of-the-art results from foundation models? - How do you adapt and extend protein and antibody models so customers can fine-tune on proprietary sequence data and get therapeutically relevant outputs? You will work across drug discovery (protein engineering, antibody design) and clinical trial operations (agentic automation, structured reasoning, domain adaptation). You will own end-to-end scientific solutions from research through production, and your work will directly shape the tools that thousands of scientists use daily. Key job responsibilities - Set the scientific vision and research agenda for LLM reasoning, agentic AI, and biological model customization across the portfolio - Innovate on LLM reasoning, planning, and agentic approaches for complex scientific and regulatory workflows - Develop model customization methods (fine-tuning, RLHF, retrieval augmentation, domain adaptation) that enable customers to train better models on their own data with less effort - Advance methods to adapt and extend biological foundation models for customer-specific therapeutic applications - Solve open research problems in faithful reasoning, multi-step planning, and tool use in high-stakes scientific domains - Partner with Life Sciences domain experts and customers to understand their hardest scientific challenges and translate those into tractable research problems - Publish at top-tier venues and build the team's external scientific reputation - Mentor applied scientists across the team while maintaining significant personal research contribution - Collaborate with product and engineering to ensure research translates into shipped products that serve customers at scale - Influence multi-year research roadmaps through deep scientific expertise and customer understanding A day in the life - Push a new reasoning approach into production that measurably improves outputs for a pharma customer's workflow - Design and run experiments to validate a novel fine-tuning method, then ship it as a capability customers can use immediately - Unblock a delivery milestone by diagnosing why a model is failing on a new class of inputs and implementing a fix - Meet with a customer's scientific team to scope what the next model release needs to do for them - Review a teammate's experimental results, sharpen the approach, and help get it over the finish line - Publish results from shipped work at a top venue, closing the loop between research and impact - Prototype a new idea that could become the next major capability in the product
US, WA, Seattle
We are seeking a Senior Manager, Applied Science to build and lead the science organization across Agentic WorkSpaces. This is a foundational leadership role spanning the full portfolio — Personal, Applications, and Core, and the agentic surfaces (WS4Builders and WorkSpaces for Agents). You will hire, grow, and lead a team of applied scientists who define how we measure and improve the performance of AI agents and human-AI teams. A core part of the role is defining the science agenda itself — identifying which problems are most worth solving and where the highest-leverage bets lie. Directions worth exploring might include Organizational Intelligence (turning institutional knowledge into agent-consumable skills), AI Agent Experience / AiAX (agent observability and autonomous remediation), and contextual, behavioral security that adapts enforcement in real time for human and agent sessions — but these are illustrative examples, not a fixed roadmap, and many other directions are possible. You and your team will define which ones we pursue. The problems your team will solve do not have established industry patterns. You will set the scientific direction and build the team that determines how AI agents and people perceive, reason about, and act reliably within computing environments at enterprise scale. What You Will Do Build and lead the applied science team. Hire, develop, and retain a high-caliber team of applied scientists spanning the Agentic WorkSpaces portfolio. Set the bar for scientific talent, create the growth paths, and build the culture that makes AAWS a destination for the best agent and human-AI researchers. Own the science strategy across the portfolio. Direct the research agenda for how we measure and improve agents and human-AI teams: the benchmarks, task suites, and metrics (accuracy, cost-per-task, task completion, productivity) that turn subjective "it works" judgments into rigorous, reproducible measurement that gates what we ship. Define and drive high-leverage research directions. Work with your team to identify the problems most worth solving and shape the science agenda. Directions worth exploring might include how agents combine deterministic tool use (MCP) with visual reasoning from computer use; Organizational Intelligence and workflow learning (learning from expert recordings, voice annotations, and SOPs); and AI Agent Experience / AiAX (detecting when agents are stuck or degrading productivity and autonomously remediating) — these are illustrative starting points, and your team will weigh them against many other possibilities. Translate science into shipped product. Partner with engineering, product, and program leaders to move models, evaluation, and learning systems from prototype into a decade-old production service operating at massive scale, without compromising the reliability that customers depend on. Represent science in leadership and to customers. Be the scientific voice in org-level planning and roadmap decisions across AAWS, and engage directly with enterprise customers on how agent performance, safety, and human-AI productivity are measured and earned. Key job responsibilities Build and lead the applied science team. Hire, develop, and retain a high-caliber team of applied scientists spanning the Agentic WorkSpaces portfolio. Set the bar for scientific talent, create the growth paths, and build the culture that makes AAWS a destination for the best agent and human-AI researchers. Own the science strategy across the portfolio. Direct the research agenda for how we measure and improve agents and human-AI teams: the benchmarks, task suites, and metrics (accuracy, cost-per-task, task completion, productivity) that turn subjective "it works" judgments into rigorous, reproducible measurement that gates what we ship. Define and drive high-leverage research directions. Work with your team to identify the problems most worth solving and shape the science agenda. Directions worth exploring might include how agents combine deterministic tool use (MCP) with visual reasoning from computer use; Organizational Intelligence and workflow learning (learning from expert recordings, voice annotations, and SOPs); and AI Agent Experience / AiAX (detecting when agents are stuck or degrading productivity and autonomously remediating) — these are illustrative starting points, and your team will weigh them against many other possibilities. Translate science into shipped product. Partner with engineering, product, and program leaders to move models, evaluation, and learning systems from prototype into a decade-old production service operating at massive scale, without compromising the reliability that customers depend on. Represent science in leadership and to customers. Be the scientific voice in org-level planning and roadmap decisions across AAWS, and engage directly with enterprise customers on how agent performance, safety, and human-AI productivity are measured and earned. Set the long-term scientific vision and team strategy: Define what best-in-class agent performance, evaluation, and learning look like across Agentic WorkSpaces — for computer-using agents and human-AI teams alike. Chart a multi-year research roadmap, and build the team and plan to deliver it. Secure buy-in from VP-level leadership. Hire and grow scientific talent: Own recruiting, calibration, development, and retention for the science team. Mentor scientists toward senior and principal scope, and raise the scientific bar across the organization. Direct research on highly ambiguous, novel problems: Guide the team through foundational challenges in agent perception, reasoning, evaluation, reliability, and human-AI collaboration — problems where neither the approach nor the success criteria are pre-defined. Drive cross-organizational alignment: Work across partner teams (AgentCore, Bedrock model teams, Identity, Security, the MCP ecosystem) and across the Applied AI Solutions product portfolio, with product and engineering leadership, to ensure scientific decisions compose into a coherent product. Deliver measurable business impact: Ensure your team's research translates to customer outcomes: higher task accuracy, lower cost-per-action, faster time-to-production, measurable productivity for human-AI teams, and the trust that lets enterprises scale agent workflows. Establish scientific rigor and operational excellence: Set the standard for experimentation, evaluation, and reproducibility, and the mechanisms that keep the science organization productive and accountable. Advance the state of the art: Enable and champion contributions to the external technical community through publications, patents, and open-source work that position AWS as the leader in the science of secure agent-computer interaction and human-AI teamwork. About the team AWS Applied AI Solutions' (AAIS) vision is every business innovating with Amazon AI teammates. Our mission is to build delightful AI solutions that improve human capabilities and business outcomes. The Agentic WorkSpaces organization within AAIS envisions a world where people, teams, and AI collaborate securely from anywhere to create unprecedented value for every organization. We build lovable products that empower every business to unlock the full potential of human-AI teamwork, driving smarter decisions, greater creativity, more value, and faster innovation with confidence. Amazon Agentic WorkSpaces (AAWS) is building the world's most lovable, secure, and trusted always-on workspace where AI agents and humans work as partners behind enterprise-grade security. Our portfolio spans persistent desktops (Personal), application streaming (Applications), and Core, and is evolving into the governed operating environment for the hybrid workforce: humans get AI-native desktops for their role, and agents get governed desktops scoped to their task, with administrators managing both as one. This surface includes WS4Builders (an AI-native environment for builders) and WorkSpaces for Agents (W4A) — enabling AI agents to work the way humans do, with access to real applications, real interfaces, and real computing environments. Enterprises want to use AI agents for critical business workloads that touch legacy desktop applications and mainframes, yet 75% of organizations run legacy applications that lack modern APIs, and 90% of corporate data remains locked in systems never designed for agents. Agentic WorkSpaces solves this: it gives enterprises a secure, governed environment where agents and humans operate both legacy and modern applications directly, just as an employee would, without costly migrations.
IN, HR, Gurugram
Work on ML teams building large-scale forecasting and optimization systems that power Amazon’s global transportation network and directly impact customer experience and cost. As an Applied Scientist II, you will set scientific direction, mentor applied scientists, and partner with engineering and product leaders to deliver production-grade ML solutions at massive scale. Key job responsibilities 1. Lead and grow a high-performing team of Applied Scientists, providing technical guidance, mentorship, and career development. 2. Define and own the scientific vision and roadmap for ML solutions powering large-scale transportation planning and execution. 3. Guide model and system design across a range of techniques, including tree-based models, deep learning (LSTMs, transformers), LLMs, and reinforcement learning. 4. Ensure models are production-ready, scalable, and robust through close partnership with stakeholders. Partner with Product, Operations, and Engineering leaders to enable proactive decision-making and corrective actions. 5. Own end-to-end business metrics, directly influencing customer experience, cost optimization, and network reliability. 6. Help contribute to the broader ML community through publications, conference submissions, and internal knowledge sharing. A day in the life Your day includes reviewing model performance and business metrics, guiding technical design and experimentation, mentoring scientists, and driving roadmap execution. You’ll balance near-term delivery with long-term innovation while ensuring solutions are robust, interpretable, and scalable. Ultimately, your work helps improve delivery reliability, reduce costs, and enhance the customer experience at massive scale.
US, NY, New York
At Amazon Selection and Catalog Systems (ASCS), our mission is to power the online buying experience for customers worldwide so they can find, discover, and buy any product they want. We innovate on behalf of our customers to infer relationships between products in Amazon Catalog to drive the selection gateway for the search and browse experiences on the website. We're solving a fundamental AI challenge: establishing product identity and relationships at unprecedented scale. Using Generative AI, Visual Language Models (VLMs), and multimodal reasoning, we determine what makes each product unique and how products relate to one another across Amazon's catalog. The scale is staggering: billions of products, petabytes of multimodal data, millions of sellers, dozens of languages, and infinite product diversity—from electronics to groceries to digital content. The research challenges are immense. GenAI and VLMs hold transformative promise for catalog understanding, but we operate where traditional methods fail: ambiguous problem spaces, incomplete and noisy data, inherent uncertainty, reasoning across both images and textual data, and explaining decisions at scale. Establishing product identities and groupings requires sophisticated models that reason across text, images, and structured data—while maintaining accuracy and trust for high-stakes business decisions affecting millions of customers daily. Amazon's Item and Relationship Platform group is looking for an innovative and customer-focused applied scientist to help us make the world's best product catalog even better. In this role, you will partner with technology and business leaders to build new state-of-the-art algorithms, models, and services to infer product-to-product relationships that matter to our customers. You will pioneer advanced GenAI solutions that power next-generation agentic shopping experiences, working in a collaborative environment where you can experiment with massive data from the world's largest product catalog, tackle problems at the frontier of AI research, rapidly implement and deploy your algorithmic ideas at scale, across millions of customers. Key job responsibilities * Formulate novel research problems at the intersection of GenAI, multimodal learning, and large-scale information retrieval—translating ambiguous business challenges into tractable scientific frameworks * Design and implement leading models leveraging VLMs, foundation models, and agentic architectures to solve product identity, relationship inference, and catalog understanding at billion-product scale * Pioneer explainable AI methodologies that balance model performance with scalability requirements for production systems impacting millions of daily customer decisions * Own end-to-end ML pipelines from research ideation to production deployment—processing petabytes of multimodal data with rigorous evaluation frameworks * Define research roadmaps aligned with business priorities, balancing foundational research with incremental product improvements * Mentor peer scientists and engineers on advanced ML techniques, experimental design, and scientific rigor—building organizational capability in GenAI and multimodal AI * Represent the team in the broader science community—publishing findings, delivering tech talks, and staying at the forefront of GenAI, VLM, and agentic system research
US, WA, Seattle
Trusted by more startups around the world, AWS makes the power of cloud computing accessible for all by giving founders everywhere access to the same technology that powers the world's largest companies. With nearly two decades of experience supporting hundreds of thousands of startups, including 80% of unicorns, we democratize cloud computing to help founders bring their innovative ideas to life. We support founders at every stage of their journey, from initial onboarding and credit programs to AI-powered guidance and scale solutions. Data is central to how we do this: it helps us identify high-potential startups early, personalize the guidance we deliver, and prioritize where we can create the most value for founders and for AWS. We are seeking an Applied Science Manager to lead a team of applied scientists and analysts building the data and machine learning capabilities behind AWS Startups. You will own the science roadmap end-to-end, from the data foundation that unifies signals about founders, startups, and their products, through a portfolio of machine learning models, to the surfaces that put insights in the hands of the teams and products that serve startups. You will balance hands-on technical leadership with people management, setting the technical bar for your team while developing their careers. Key job responsibilities · Lead, coach, and grow a team of applied scientists, business intelligence engineers, and business analysts; hire and develop talent and set a high technical bar. · Own and prioritize the team's science roadmap and set technical direction for its machine learning models and data assets, balancing rapid experimentation with production quality, cost, and reliability. · Scope scientific projects, design and evaluate experiments, and productionize models that deliver measurable impact, establishing measurement, evaluation, and operational-excellence standards so quality and impact are quantified and defensible. · Drive the science behind recommendation systems, startup segmentation and targeting, and fraud detection, delivering models that surface relevant opportunities, group and prioritize startups by need and fit, and protect the business from fraud and abuse. · Partner with product, engineering, design, and go-to-market teams to translate science into scalable products, and communicate strategy, results, and trade-offs clearly to technical and non-technical leaders. · Foster a culture of scientific rigor and rapid experimentation, and proactively identify and escalate risks with clear mitigation plans. About the team The AWS Startups team builds innovative products and platforms that support startup customers throughout their journey, from initial onboarding and credit programs to AI-powered guidance and scale solutions. Our portfolio serves hundreds of thousands of startup customers globally, and we partner with business development, field marketing, and solutions architecture teams worldwide. We are building the next generation of AI-native products that make world-class cloud expertise accessible to every founder.
IN, KA, Bengaluru
Amazon Ads is a multi-billion dollar global business that delivers advertising experiences across Amazon's owned-and-operated properties (including Prime Video, Twitch, Fire TV, and Amazon.com), third-party publisher networks, and emerging channels like generative AI-powered shopping experiences. As one of the fastest-growing segments of Amazon, we operate at unprecedented scale across desktop, mobile, connected TV, and emerging surfaces. Within Amazon Ads, Traffic Quality is a critical pillar of advertiser trust and marketplace integrity. Our mission is to build advanced capabilities that work at petabyte scale to detect sophisticated invalid traffic (IVT) which includes sophisticated non-human traffic, bot networks, and fraudulent engagement patterns across programmatic advertising. We are on a journey to establish Amazon Ads as an industry leader in traffic quality standards and transparency. Our research agenda focuses on staying ahead of adversarial actors through continuous innovation in detection methodologies, leveraging state-of-the-art techniques in deep learning and generative modeling, user behavior and multi-modal representation learning, anomaly detection, time-series analysis, and sparse labeling methods. We process billions of ad events daily, developing novel algorithms that balance precision and recall while operating under strict latency constraints. Our work directly protects hundreds of millions of dollars in advertiser spend annually while maintaining a seamless user experience. Key job responsibilities As a Data Scientist II in Traffic Quality, you will solve inherently hard problems in advertising fraud detection by applying advanced statistical techniques and machine learning. You'll work on systems that process billions of ad impressions and clicks per day, using Amazon's cloud services including EC2, S3, EMR, Sagemaker, and RedShift. - Define and frame new research problems in fraud detection where neither problem nor solution is well-defined. - Apply new machine learning approaches, models, and algorithms to detect sophisticated invalid traffic. - Apply domain knowledge to perform broad data analysis as a precursor to modeling and build business insights. - Work with unstructured and massive datasets to deliver results. - Produce research reports meeting top-tier external publication standards. - Mentor and develop junior scientists on the team. About the team Here are a few papers published by the team: 1/ [Scaling Generative Pre-training for User Ad Activity Sequences. AdKDD 2023.](https://assets.amazon.science/b7/42/03be071743d5a57cb1656e6caa34/scaling-generative-pre-training-for-user-ad-activity-sequences.pdf) 2/ [SLIDR: Real-time Robot Detection On Online Ads, IAAI 2023, Deployed Highly Innovative Applications of AI Track (AAAI 2023)](https://assets.amazon.science/75/2f/3b7106b143f38f7f4d2806388ace/real-time-detection-of-robotic-traffic-in-online-advertising.pdf) 3/ [Self-supervised Representation Learning Across Sequential and Tabular Features Using Transformers, NeurIPS 2022, First Table Representation Learning Workshop](https://openreview.net/forum?id=wIIJlmr1Dsk)
US, WA, Seattle
Pricing is one of the most consequential decisions Amazon makes — and the science behind it needs to be causally rigorous, not just predictive. The P2 Optimization Science (P2OS) team builds the machine learning systems that power Amazon's pricing decisions at scale: demand lift models, customer lifetime value frameworks, and the experimentation infrastructure that validates whether our pricing changes actually work. We're hiring an Applied Scientist to own causal inference at the intersection of ML and pricing experimentation. This role exists because our team has identified a real gap: the methodological bridge between econometric analysis (owned by our economists) and production-scale ML pipelines (owned by our engineers) needs a practitioner who lives in both worlds. You'll build CATE estimation models, design analysis workflows for pricing weblabs, and develop the reusable causal ML infrastructure that the broader team — including non-ML scientists — can rely on. This is not a research role. The bias here is toward shipping production-quality causal pipelines with real downstream business impact. You'll measure success by what changes in LTV estimates, what pricing errors your models help avoid, and whether the economists on your team can actually use what you build. If you're a scientist who wants to work on hard causal identification problems in a high-stakes production environment — and who finds satisfaction in making rigorous methods accessible to a broader team — this role is for you. Key job responsibilities * Build causal ML pipelines for pricing — Design, train, evaluate, and deploy end-to-end causal estimation models for pricing use cases. * Own the science on heterogeneous treatment effects — Be the team SME on causal ML methodology: identification strategies, model selection, evaluation standards, and the tradeoffs between econometric and ML approaches to causal estimation. * Support pricing experiment analysis — Contribute causal analysis methodology to pricing weblab and A/B test post-analysis; build reusable tooling that economists can use without requiring ML expertise * Connect model outputs to business outcomes — Define, before writing code, what business metric each model moves; deliver model evaluation reports framed around pricing errors avoided and LTV estimate changes. * Evaluate and adopt novel techniques — Assess applicability of emerging causal inference methods (synthetic DiD, generalized random forests, causal representation learning) to Amazon's pricing context; write internal methodology proposals for adoption * Write internal documentation and methodology papers — Produce at least one internal write-up per half that connects a causal ML technique to a concrete pricing use case; make pipelines extensible and well-documented so other scientists can build on them. * Collaborate across disciplines — Partner closely with the Sr. Economist on identification strategy and causal assumptions; work with SDE and DE partners on production deployment; align with PMs on experiment design requirements A day in the life As an Applied Scientist on the P2OS team, your work directly shapes the prices customers see on hundreds of millions of Amazon products. In a given workweek, you might: * Investigate an optimization anomaly in simulation and trace it back to a model input gap or an unmodeled market dynamic * Design an offline evaluation framework to benchmark competing optimization approaches before committing to online testing * Collaborate with Sr. Economists on the identification strategy for the model you're building for a pricing lab * Present a science proposal for incorporating a new competitiveness or inventory signal into an optimization system * Work cross-team with the experimentation platform team on randomization design. * Develop and write up a novel scientific finding — preparing a paper or technical report for submission to a top-tier venue such as KDD, NeurIPS, or the ACM Conference on Economics and Computation
US, CA, Sunnyvale
We are seeking an Applied Scientist to focus on Robot Navigation. In this role, you'll research and develop advanced navigation systems that enable robots to move reliably and safely through complex, dynamic environments. You'll work across a broad spectrum of navigation approaches—from classical methods to learning-based techniques and foundation models—to build robust solutions for autonomous robot navigation. Key job responsibilities - Develop and implement robust navigation systems that enable reliable autonomous operation in complex, dynamic indoor environments with static and dynamic obstacles - Build simulation-based and on-device evaluation frameworks with comprehensive benchmarks and metrics for systematic comparison of navigation methods - Conduct sim-to-real transfer experiments, analyzing performance gaps and developing techniques to ensure reliable real-world navigation performance - Collaborate with world model, manipulation, and other teams to ensure seamless integration of navigation capabilities into the full robot system - Stay current with the latest advances in robot navigation, spatial reasoning, and related fields, and apply relevant findings to improve system performance - Mentor fellow scientists and engineers while maintaining strong individual technical contributions About the team Fauna Robotics, an Amazon company, is building capable, safe, and genuinely delightful robots for everyday life. Our goal is simple: make robots people actually want to live and interact with in everyday human spaces. We believe that future won’t arrive until building for robotics becomes far more accessible. Today, too much effort is spent reinventing the fundamentals. We’re changing that by developing tightly integrated hardware and software systems that make it faster, safer, and more intuitive to create real-world robotic products.
IN, KA, Bengaluru
The Ads Trust Science team, based in Bangalore, is responsible for ensuring that ads are relevant and is of good quality, leading to higher conversion for the sellers and providing a great experience for the customers. We deal with one of the world’s largest product catalog, handle billions of requests a day with plans to grow it by order of magnitude and use automated systems to validate tens of millions of offers submitted by thousands of merchants in multiple countries and languages. In this role, you will build and develop ML models to address content understanding problems in Ads. These models will rely on a variety of visual and textual features requiring expertise in both domains. These models need to scale to multiple languages and countries. You will collaborate with engineers and other scientists to build, train and deploy these models. As part of these activities, you will develop production level code that enables moderation of millions of ads submitted each day.
PL, Gdansk
Have you ever wondered how we give voice to devices — even when they're offline? The Text-to-Speech on Device team at Amazon builds AI-powered voice models that run locally on hardware with limited resources, serving customers across Alexa, automotive, and accessibility experiences for visually impaired users. We sit at the intersection of speech generation, generative AI, and on-device machine learning, and we're looking for a curious, collaborative Applied Scientist to help us push what's possible. In this role, you will research and develop production-ready speech generation models optimized for constrained environments. You will work across the full model lifecycle — from early experimentation and prototyping through to integration on real devices. If you're excited about solving hard scientific problems that directly improve how millions of people interact with technology, we'd love to hear from you. Key job responsibilities - Design and develop end-to-end machine learning models for on-device speech generation, from early research and experimentation through production-ready deployment. - Research and apply advanced techniques in generative AI, model compression, and knowledge distillation to deliver high-quality voice models within tight hardware constraints. - Propose and validate novel scientific approaches by authoring detailed technical specifications and contributing to peer-reviewed publications when appropriate. - Evaluate model performance rigorously, identify improvement opportunities, and iterate on training and inference pipelines to optimize quality and efficiency. - Collaborate with science and engineering teams across cloud and device platforms to bring speech generation capabilities from research prototypes to integrated product experiences. About the team The Text-to-Speech on Device team builds low-footprint AI models for speech generation that run locally on devices such as Android and FireOS platforms. Our models require significantly less computation than cloud-hosted alternatives, enabling offline voice experiences for Alexa, automotive partners, and accessibility solutions. We work closely with device engineering teams and cloud-based speech science teams to deliver the best possible experience for our customers. Our focus in the coming years is expanding the range of voices and languages we support while continuing to improve naturalness and efficiency on constrained hardware.