In short
Founding engineer (employee #4) at an early-stage London startup building a searchable memory for private equity and investment firms: reading documents and emails with LLMs, building knowledge graphs, and the retrieval and agents on top. Stack is Python, TypeScript and Google Cloud; hybrid in London, £50K–£100K plus founding equity.
About the role
Founding Engineer - AI startup for private equity - London (hybrid)
£50K - £100K + Equity
Private equity firms lose knowledge constantly. Years of deals, memos and opinions are buried in inboxes and folders, and a lot of it leaves when a partner does.
I’m working with an early-stage London startup that fixes this. They’re building a searchable memory for investment firms. It reads a firm’s documents and emails, then connects everything it finds into one record, so you can ask a question and get an answer with the source attached.
They need a founding engineer, Employee #4.
You’d take a problem from a pile of messy PDFs to an answer a partner will bet money on, and you’d own it the whole way through. That covers reading documents with LLMs, building the knowledge graph, and the retrieval and agents on top. Python, TypeScript, Google Cloud.
It’s a tiny team run by second-time founders, with real say in how the platform gets built and founding equity to go with it.
They want someone who has shipped retrieval, extraction, knowledge graph or infrastructure work in production and wants to own it.
Requirements
- Has shipped retrieval, extraction, knowledge graph or infrastructure work in production
- Build document-reading pipelines with LLMs, plus knowledge graphs, retrieval and agents
- Python
- TypeScript
- Google Cloud
- Comfortable owning problems end-to-end, from messy PDFs to an answer a partner will bet money on