AI Engineer

AI Engineer (Applied Agents)

Location: London (Hoxton) (4-5 Days In-Office)

Founders: Ex-Palantir founders

Stage: Seed / Series A

The Opportunity

This founding team, led by a ex-Palantir vet, who is building a platform to extract "institutional soul" from unstructured data.

They are moving beyond simple chatbots to create a reasoning engine for the enterprise. Their platform ingests the fragmented reality of Slack, emails, and tickets to automate high-stakes workflows that previously required manual, human-intensive research. They are currently operating with high urgency and a "production-first" mindset to redefine how AI learns business principles.

Example projects you would work on:

• Build the systems that navigate thousands of fragmented emails to generate high-fidelity, first-draft responses for customer-facing teams—augmenting their ability to handle complex queries without hours of manual data retrieval.
• Develop agents that don't just "chat," but navigate complex enterprise databases to solve operational disputes autonomously.
• Build automated pipelines using synthetic data and human feedback to measure the nuance, tone, and factual accuracy of agentic decision-making.
• Implement logic that allows an LLM to detect its own hallucinations in a workflow and force a re-plan based on ingested technical documentation.
• Shave RAG query times down from "noticeable lag" to "instant" across massive, unstructured enterprise datasets.

Why a Senior Engineer would join this team

• You’ll work directly with founders who have already scaled AI in the most demanding environments on earth (defense, government, and global finance).
• This isn't a lab; it’s an applied AI company. You’ll be dealing with the "messy" reality of unstructured data, which is where the real value—and the highest technical barrier—lies.
• The team operates without PMs. Engineers own the product end-to-end, from the first line of Python to the customer outcome.
• You are joining a category-defining company in its earliest, most impactful stage. The equity upside and technical influence reflect that.
• They believe the hardest problems are solved in person. This is an environment for people who want to move fast, collaborate at the whiteboard, and ship daily.

Who they are looking for

• You have deep, non-generic expertise in LLMs and RAG. You didn’t just follow a tutorial; you’ve seen these systems fail in production and know how to make them robust.
• You prefer shipping code that works over writing papers. You value the "messy" side of data and enjoy the challenge of making AI work in unpredictable environments.
• You can explain complex multi-step reasoning to a non-technical stakeholder clearly, then dive straight into the FastAPI or Next.js code to build it.
• This team thinks about the product at night. They are looking for a collaborator who is equally obsessed with the potential of agentic systems.

The Technical Stack

• Backend: Python (FastAPI)
• Frontend: TypeScript (Next.js)
• AI: LLM Orchestration, Agentic Frameworks, Vector DBs
• Infrastructure: AWS

The Interview Process

1. Intro Call (30-min): A high-level alignment on vision and technical philosophy with the founders.
2. Technical Deep Dive (60-min): A walkthrough of real systems you’ve built. No generic brain teasers—only architecture and code.
3. Paid Work Day: A specific, async technical challenge based on a real problem the team is currently solving.
4. Final Founder Chat: A deep dive into the roadmap, equity, and long-term vision.
5. Offer.