Amazon MLE Predictive Metrics Alexa Conversational Assistant Services 

United States, Massachusetts, Boston 

221199657

As our SDE/MLE superstar, you'll have the power to lead the charge in developing cutting-edge inference and architectures to ensure our scientists can push the boundaries of large model training (esp. experience with AWS Inferentia2). Your groundbreaking work will directly impact our customers' lives through life-changing and delightful products and services.
Key job responsibilities- Responsible for the development and maintenance of key platforms needed for developing, evaluating and deploying LLM for real-world applications.- Work closely with Applied scientists to process massive data (e.g., 20B parameter), scale machine learning models while optimizing.
- Hold and raise the bar for operational readiness, operational excellence and runbook documentation.
- Ability to be curious and pivot your perspective.
A day in the life
- Design, document, and build novel inference infrastructure and/or modeling techniques using hardware like AWS Inferentia2 (and/or others).
- Inform and guide predictive metrics roadmap based on learnings, opportunities, and efficiencies in hardware and infrastructure.
- Attend daily/weekly/monthly/quarterly scrum activities to ensure technical delivery is on-time, under-budget, and meeting customer requirements.
- Hold and raise a high bar for technical implementation standards -- from code revision submissions/reviews to operational readiness and excellence.

- 3+ years of non-internship professional software development experience
- 2+ years of non-internship design or architecture (design patterns, reliability and scaling) of new and existing systems experience
- Experience programming with at least one software programming language

- 3+ years of full software development life cycle, including coding standards, code reviews, source control management, build processes, testing, and operations experience
- Bachelor's degree in computer science or equivalent
- Expertise in model deployment on inf2