AI Engineer
Nava Benefits · Remote · Remote ·
COMPENSATION
€•• – ••k
Listed on the employer's job board (Ashby)
- Fully remote
- AI Platform Engineering
- TypeScript + Python
- LLM Infrastructure
- Senior Level
You will transform initial prototypes into a robust platform that supports all AI teams at Nava. This involves creating safe execution environments for long-running agents and optimizing model usage to reduce costs and engineering overhead.
What you'll do
- Harden initial capabilities like context management and agent state for reuse across products
- Design state management patterns so long-running agent tasks survive deploys and restarts
- Build mechanisms for packaging context into zero-trust execution environments
- Optimize model routing and subagent orchestration to reduce cost per task at scale
- Reduce unplanned platform work for product engineers by providing stable shared patterns
- Document, test, and trace shared patterns to ensure maintainability and quick onboarding
What you'll need
- Experience owning infrastructure for LLM products, such as agent runtime or orchestration
- Proven track record of shipping LLM features to production and managing tradeoffs
- Proficiency in modern full-stack development using TypeScript and/or Python
- Experience integrating foundation-model APIs, tool calling, and human review flows
- Practical habits for evaluating and tracing model behavior in real-world use cases
- Ability to translate problems from product and design experts into reusable technical capabilities
Stack
- React
- Node
- TypeScript
- Python
- Postgres
- AWS
- Langfuse
How you'll work
- Remote-first company culture
- Small, senior team structure
- Direct access to leadership for decision making
What you get
- Protected space to build core AI platform
- Work on frontier problems with real healthcare users
- Inclusive and diverse team environment
Hiring process
- 15-minute introduction with recruiter
- 15-minute AI conversation about notable LLM work
- 60-minute live coding session in your usual environment
- 60-minute design conversation
- 60-minute accomplishment discussion