Multi-Agent LLM Systems
Remote (MUST BE IN EUROPE)
We’re partnering with a venture-backed startup led by a founder who has built and taken two technology companies to IPO, now assembling a world-class team to tackle one of the most impactful problems in applied AI.
The company is developing a voice-enabled AI copilot used by professionals to eliminate the friction from documentation and decision-making, a product with genuine, real-world impact that’s already being used in production environments.
They’re now looking for a Senior/Staff AI Engineer to own and evolve the core “brain” service behind this assistant, the system that powers reasoning, retrieval, and dialogue in real time.
Interview Process:
1️⃣ Intro call where we talk about the role.
2️⃣ Technical discussion with the Head of AI.
3️⃣ Deep-dive session with a Backend Engineer and ML Engineer from the team.
4️⃣ 30-minute conversation with theFounder.
Why This Is Worth Your Time
- Real ownership: You’ll be the architect behind a core AI system, not a feature contributor.
- Fast-moving environment
- Immediate impact: Your code will run in production and support real users from day one.
- Technical depth: Multi-agent reasoning, voice-streaming, RAG optimisation and all in one system.
- Flexible setup: Remote across the EU, with optional co-working in London or Barcelona.
What you’ll do
- Obsessive about latency, you think in milliseconds, optimise for concurrency, and understand the trade-offs between speed, cost, and model performance.
- Design, implement, and productionise multi-agent LLM systems that reason, plan, and coordinate.
- Develop FastAPI-based microservices optimised for low latency and high reliability.
- Engineer and evaluate RAG pipelines: hybrid retrieval, re-ranking, grounding, and context validation.
- Integrate real-time voice interfaces (STT/TTS, WebRTC, LiveKit) into intelligent conversational flows.
- Instrument and evaluate system performance using observability and model-faithfulness metrics.
What we’re looking for
- Proven ability to build and ship agentic or multi-agent frameworks into production.
- Expert Python, FastAPI, and asyncio developer.
- Practical experience with LangChain, Autogen, or custom orchestration layers.
- Startup mindset: ownership, speed, and pragmatism over perfection.
Bonus points
- Experience working with voice or streaming systems (STT/TTS, WebRTC, LiveKit).
- Exposure to evaluation tooling, LLM-as-judge setups, or agent benchmarking.
- Background in healthtech, fintech, or other compliance-heavy sectors.
