Principal AI Engineer, Agentic Conversational Platform

Socket.dev; Bellevue (WA); On-site; USD 200,000 - 271,000; Full time

Benefits offered by this job

RSUs

Comprehensive benefits

Job summary

Amazon is building Alexa AI’s agentic platform, overseeing the end-to-end lifecycle from evaluation to training and serving for LLM-based agents. You will define and own the architecture for scalable, trustworthy, production-grade systems that can handle long-running, multi-turn interactions at high concurrency.

You will partner with scientists and other engineers to set the technical bar, drive design reviews, and mentor senior engineers, delivering platforms that scale and improve through

Qualifications

• 12+ years of non-internship professional software development experience.
• Knowledge of object-oriented design, data structures, and algorithms.

Responsibilities

• Define and drive the engineering roadmap and architecture for the agentic AI platform: evaluation, training, self-learning, and serving for LLM-based agents in production
• Architect large-scale agentic evaluation infrastructure with sandboxed execution, recreatable environments, verifiable scoring, and reproducibility at hundreds of concurrent trials
• Build and scale RL and post-training systems for agentic workloads: 256K+ token contexts, multi-turn trajectory training, train/inference engine consistency
• Design the serving and inference architecture for agentic traffic, KV-cache optimization, and cross-team collaboration
• Set the technical bar across the organization through design reviews, operational excellence, and cross-team problem solving
• Translate product/science requirements into platform interfaces and influence senior leadership on build-vs-adopt and ownership decisions
• Mentor and grow senior and principal-track engineers across multiple teams

Skills

OO design

Data structures

Algorithms

Job description

Amazon is building Alexa AI’s agentic platform, overseeing the end-to-end lifecycle from evaluation to training and serving for LLM-based agents. You will define and own the architecture for scalable, trustworthy, production-grade systems that can handle long-running, multi-turn interactions at high concurrency.

You will partner with scientists and other engineers to set the technical bar, drive design reviews, and mentor senior engineers, delivering platforms that scale and improve through