Principal AI Software Engineer
Skills
APIAmazon Web ServicesEngineeringGithubMicrosoft SurfaceMulti-Tenant ManagementReasoning Skills
What the job involves
The main requirements, responsibilities and hiring steps.
Requirements
- 10+ years designing complex distributed systems with production intuition from delivering complete end-to-end software in real environments
- AI-native across the full development lifecycle using tools like Cursor Claude Code GitHub Copilot and OpenAI Codex
- Fluent across the full AI model arc in production and able to choose the right approach for each use case
- Built real agentic systems with failure modes recovery and human-in-the-loop control
- Able to solve across the company not just within one function and energized by an early-stage environment
- Ship fully working use cases with multi-tenant and access-controlled design from day one
Nice to have
- AI-native
- Full-spectrum ownership
- Early-stage mindset
- Systems thinker
- Cross-functional
Day to day
- Own agentic AI systems and MCP server development, including autonomous task orchestration, tool registries, multi-step reasoning, and human-in-the-loop control.
- Build voice AI and conversational systems for real operational environments with real-time audio pipelines, production-grade latency, reliability, and concurrent session handling.
- Lead full-stack platform engineering across Next.js and TypeScript frontend, AWS Lambda and serverless backend, DynamoDB, SQS, and streaming APIs, delivering complete multi-tenant and access-controlled solutions end to end.
- Drive the full AI model lifecycle from prompt engineering and RAG to fine-tuning and dedicated model training for text and image, while creating operational data pipelines and self-improving systems.