Artificial Intelligence Engineer

2 to 5 years

• Design, build, and deploy AI agents and LLM-powered features — prompt/harness design, tool orchestration, structured outputs.
• Integrate AI systems with third-party platforms — CRMs, messaging channels, support tools — via REST APIs, OAuth, and webhooks.
• Build and maintain reusable components (prompt libraries, guardrails, validation logic) instead of one-off scripts per use case.
• Write typed tool contracts and enforce output schemas (Pydantic or equivalent) to keep agent behavior predictable and testable.
• Set up and run evals — trajectory evals, regression tests — to catch quality drops before they reach production.
• Work with Product/Engineering to turn requirements into working agent behavior, and document architecture clearly enough for handoff.

• Experience: 2–5 years building production software, with meaningful hands-on LLM work
• Strong Python — comfortable with FastAPI or Flask for backend services.
• Practical experience with LLM APIs (Anthropic, OpenAI, or open-source models) — prompt engineering, function calling/tool use, context management.
• Working knowledge of API integration patterns — OAuth 2.0, webhooks, REST, sync vs. async/event-driven design.
• Familiarity with SQL and at least one vector/embedding store.
• Version control (Git) and basic CI/CD exposure.
• Clean, process-oriented coding habits — this is client-facing infra, not a hackathon.
• Good communication - able to explain technical trade-offs to non-technical stakeholders.

• Experience with agentic coding tools (Claude Code, Cursor) in your own workflow.
• Exposure to voice agents, multi-channel conversational AI, or real-time streaming systems.
• Experience with self-hosted auth/OAuth tooling (Nango or similar) over managed platforms.
• Docker/Kubernetes and basic cloud deployment (AWS/GCP/Azure) experience.
• Startup background — comfortable owning ambiguous problems without a fully specced ticket.