AI Engineer

Role

LLM Workflow Builder – design and wire up prompts, tools, memory and agents with LangChain / LangGraph / LangServe.

API & Micro-Service Owner – ship FastAPI endpoints (REST + streaming) that are secure, observable and lightning fast.

End-to-End Engineer – own the full lifecycle : code → container → CI / CD → K8s → monitoring, with a bias for rapid, iterative releases.

Cross-Functional Partner – collaborate with product and design to scope MVPs, estimate effort, and keep delivery on track.

Responsibilities

Architect, implement and maintain multi-agent workflows and retrieval-augmented pipelines.

Write clean, typed, tested Python (async, pytest, poetry).

Instrument and evaluate everything with LangSmith / LangFuse; build guardrails and regression suites for prompts.

Containerize (Docker) and deploy (Kubernetes, GitHub Actions) with automated rollbacks and alerts.

Optimize latency, throughput and cost; troubleshoot production issues end-to-end.

Mentor teammates on best practices in LLM app engineering and DevOps.

Must-Have Skills

Strong Python craftsmanship and async I / O expertise.

Proven experience with LangChain or LangGraph in real projects.

FastAPI (or similar) for building REST / streaming APIs.

Solid DevOps : Docker, basic Kubernetes, CI / CD pipelines, observability tooling.

Git proficiency : PR hygiene, code reviews, and clear documentation.

Nice-to-Haves

Lightweight fine-tuning (LoRA / QLoRA) or embedding-model selection.

Vector databases (Postgres / pgvector, Milvus, LanceDB).

Serverless / edge deploys (Fly.io, Cloudflare Workers, AWS Lambda).

Experience with multi-agent planning frameworks or autonomous tool-use research.