Staff LLM Engineer — Agents & enterprise RAG

Thames Inference Labs is hiring a Staff Staff LLM Engineer — Agents & enterprise RAG on the Applied LLM team (Hybrid · London, United Kingdom). This is a hands-on role in LLM & generative AI—not a generic “AI enthusiast” post.

You will own LLM features, retrieval, agents, and evaluation end to end: problem framing, offline evaluation, safe rollout, and iteration from production signals. Success is measured with clear metrics (quality, latency, cost, or business KPIs)—not slide decks alone.

Thames Inference Labs ships assistant and RAG platforms for regulated customers in banking, insurance, and critical infrastructure. We blend pragmatic delivery with strong evaluation and rollback discipline so models never become a black-box liability.

We are hiring a Staff LLM Engineer based in London (hybrid). You will anchor technical decisions for retrieval, tool-calling, and evaluation—partnering daily with applied scientists, security, compliance, and forward-deployed teams.

What you will do

• Architect multi-tenant RAG pipelines: ingestion, ACL-aware retrieval, citations, summarisation policies, and human review hooks where required.
• Implement agent workflows with deterministic fallbacks—tracing, cost controls, idempotent tools, and clear failure modes.
• Own offline and online evals (golden sets, regressions, calibrated LLM-as-judge) and gates before production prompt or model changes.
• Drive latency and reliability SLOs with infra partners; tune batching, caching, and serving paths for flagship workloads.
• Produce crisp RFCs and runbooks so customer-facing teams can operate flows safely after launch.

What we look for

• 7+ years backend or ML systems experience with multiple production LLM or retrieval launches.
• Expert Python and modern orchestration (LangChain/LangGraph or equivalent); deep familiarity with vector DBs and OpenAI-compatible APIs.
• Comfort owning ambiguous problem statements and communicating trade-offs to non-engineering stakeholders.
• Demonstrable mentoring and code-review habits; care for tests and observable systems.
• Eligible to work in the UK; professional English.

Nice to have

• Kubernetes, Terraform, and observability stacks (OpenTelemetry, Grafana) in regulated environments.
• Experience with EU AI Act, DPIA artefacts, or security-driven release processes.
• Conference talks, OSS, or blog posts on agents or retrieval.