Sr. Applied Scientist, SSG Science

Amazon; Bengaluru; On-site; INR 1,800,000 - 3,000,000; Full time

Job summary

Amazon Devices is an inventive R&D unit developing edge AI for Kindle, Fire, Echo, Astro and more. We seek exceptional scientists to join the Applied Science team to advance edge Gen AI models and optimization with custom ML hardware collaborations.

Ideal candidates have strong ML research background, PhD or MS with extensive applied experience, and hands-on coding in Java, C++, Python. You will work across teams to ship production edge models and contribute to open-source projects.

Qualifications

• 3+ years building ML models for business applications.
• PhD, or Master's + 4+ years of applied research.
• Proficiency in Java, C++, Python.
• Experience deploying ML models to production.

Responsibilities

• Quantize, prune, distill and finetune Gen AI models for edge platforms.
• Understand Neural Edge Engine to create optimization techniques.
• Analyze DL workloads and map them to Neural Edge Engine.
• Apply information theory, scientific computing and DL theory principles.
• Train custom Gen AI models surpassing SOTA for production.
• Collaborate with compiler engineers, hardware architects, and product teams.
• Publish in open source and present at ML conferences (NeurIPS, ICLR, MLSys).

Skills

Java

C++

Python

Deep learning

Education

PhD

Master's + 4+ years applied research

Job description

Amazon Devices is an inventive research and development company that designs and engineer high-profile devices like the Kindle family of products, Fire Tablets, Fire TV, Health Wellness, Amazon Echo & Astro products. This is an exciting opportunity to join Amazon in developing state-of-the-art techniques that bring Gen AI on edge for our consumer products. We are looking for exceptional scientists to join our Applied Science team and help develop the next generation of edge models, and optimize them while doing co-designed with custom ML HW based on a revolutionary architecture. Work hard. Have Fun. Make History.

Key job responsibilities

What will you do?

• Quantize, prune, distill, finetune Gen AI models to optimize for edge platforms
• Fundamentally understand Amazon's underlying Neural Edge Engine to invent optimization techniques
• Analyze deep learning workloads and provide guidance to map them to Amazon's Neural Edge Engine
• Use first principles of Information Theory, Scientific Computing, Deep Learning Theory, Non Equilibrium Thermodynamics
• Train custom Gen AI models that beat SOTA and paves path for developing production models
• Collaborate closely with compiler engineers, fellow Applied Scientists, Hardware Architects and product teams to build the best ML-centric solutions for our devices
• Publish in open source and present on Amazon's behalf at key ML conferences - NeurIPS, ICLR, MLSys.

Basic Qualifications

• 3+ years of building machine learning models for business application experience
• PhD, or Master's degree and 4+ years of applied research experience
• Experience programming in Java, C++, Python or related language
• Experience with neural deep learning methods and machine learning
• Prior experience in productionizing ML models and managing their full lifecycle from development to deployment

Preferred Qualifications

• Understanding and preferably hands-on experience with recent methods for inference optimization, including Mixture-of-Experts (MoE), Diffusion Models for Language Generation, etc.

Our inclusive culture empowers Amazonians to deliver the best results for our customers. If you have a disability and need a workplace accommodation or adjustment during the application and hiring process, including support for the interview or onboarding process, please visit https://amazon.jobs/content/en/how-we-hire/accommodations for more information. If the country/region you're applying in isn't listed, please contact your Recruiting Partner.

Amazon is an equal opportunity employer and does not discriminate on the basis of protected veteran status, disability, or other legally protected status.