Agentic AI Process Engineer

LifeChef is a HIPAA-compliant healthcare platform delivering Medically Tailored Meals. This role involves architecting and maintaining Enso v2, an agentic system that governs how AI builds product features through structured autonomy, validation, and human-in-the-loop oversight.

• Design agent topology for the development lifecycle including research agents, spec validation agents, code review agents, compliance checking agents, and red-team agents with defined boundaries and escalation triggers
• Build spec-to-code reconciliation systems that validate whether implemented code satisfies specifications and flag deviations
• Develop multi-model orchestration strategy routing tasks to different models based on cost, capability, and risk
• Create structured output validation ensuring agent outputs are parseable, auditable, and actionable
• Establish governance and escalation frameworks defining autonomous agent decisions versus human sign-offs
• Build evolution engines that track agent failures and feed improvements back into prompt refinement
• Integrate agentic layer with existing CI/CD, GitHub Actions, automated testing, PHI scanning, and compliance validation

• Designed or built at least one agentic system with structured autonomy, multi-step workflows, validation, error handling, and escalation
• Deep understanding of agent topology as a design problem including agent boundaries and communication patterns
• Hands-on experience with LLM orchestration frameworks such as LangChain, CrewAI, AutoGen, or custom implementations
• Experience with eval frameworks and validation methodologies including confidence scoring
• Understanding of multi-model strategies and cost/capability tradeoffs
• Proficiency in Python and TypeScript or comfort learning one while operating in the other
• Strong written communication skills for async documentation of architectural decisions
• Preferred: Model Context Protocol experience, AI-native development tools proficiency, healthcare/fintech/regulated environment experience, systems-level prompt engineering expertise, adversarial validation experience

Schedule: Full-time with 6-90 day trial sprint, heavy pairing with CTO in first 30 days with increasing autonomy

Location: Remote-friendly, small senior engineering team primarily Ukraine-based

Team Context: Small senior team, direct collaboration with CTO (Jay) who has 20+ years experience in healthcare, threat intelligence, and enterprise platforms

Tech Stack: Next.js 16, React 19, TypeScript, PostgreSQL, AWS serverless (Lambda, CDK, Amplify Gen 2), Vitest testing, Pino structured logging, HIPAA compliance