Founding Applied AI Engineer

WilsonAI Company; Greater London; On-site; GBP 80,000 - GBP 100,000; Full time

Job summary

A pioneering AI firm in the legal industry is seeking a Founding Applied AI Engineer to develop LLM-powered workflows for contract management. This hands-on role requires building reliable AI systems and diagnosing model failures. The ideal candidate has experience with production LLM workflows and strong problem-solving abilities. Join a fast-paced start-up poised for significant disruption and impact in the trillion-dollar legal sector.

Qualifications

• Experience building production LLM workflows or agent-based systems.
• Strong computer science fundamentals and clear problem solving skills.
• High agency and comfort with responsibility and speed.

Responsibilities

• Build and improve LLM driven workflows for contract review.
• Design and implement agent orchestration pipelines.
• Diagnose and fix model failures and edge cases.
• Collaborate with customers to create reliable AI systems.
• Own features from design to production.

Skills

Production LLM workflows

Deep intuition for language models

Structured problem solving

Complex systems reasoning

Attention to detail

Fast learner

Startup mentality

Job description

WilsonAI is building Cursor for lawyers. We help legal teams review, negotiate, and manage contracts with unprecedented speed and accuracy. We are backed by Nomad Ventures and Autopilot and raised a $1.7M pre seed at the end of 2024. Since then we have stabilised our product with real customers, beaten established competitors in head to head pilots with teams handling hundreds of millions in revenue, and scaled through founder led sales to multi six figure revenue. We are at the very start of a huge opportunity to transform the trillion dollar legal industry using AI, and we are building a small, elite team to do it.

The role

We are looking for a Founding Applied AI Engineer to work at the core of our product. You will work directly with our CTO across both customer facing features and deep technical systems. You will design, build, and ship LLM powered workflows that real legal teams rely on every day. You will debug failure modes, improve accuracy, and push models and systems until they work reliably in production.

This is a hands on role. You will own problems end to end, from understanding what a customer needs to shipping a working solution. You will be expected to move fast, make good decisions with incomplete information, and take responsibility for outcomes.

In return, you will get rare exposure to how a high growth AI company is built from the inside, with direct ownership over critical parts of the product.

What you will do

• Build and improve LLM driven workflows for contract review, redlining, extraction, and legal Q&A
• Design and implement agent orchestration and multi step reasoning pipelines
• Diagnose and fix model failures, hallucinations, and edge cases
• Work closely with customers and product to turn real legal problems into reliable AI systems
• Own features end to end, from design through to production and iteration

What we are looking for

• Experience building production LLM workflows or agent based systems
• Deep intuition for how language models behave and how to steer them
• Strong computer science fundamentals and clear, structured problem solving
• Ability to reason about complex systems and anticipate failure modes
• High attention to detail and a bias toward shipping things that actually work
• High agency, fast learner, and comfortable owning problems without hand holding
• Comfortable working hard and doing unglamorous work when needed
• Startup builders, ex founders, or engineers who want responsibility and speed are especially welcome

Why this role

You will be working at the frontier of applied AI in a domain where correctness and reliability really matter. The legal industry is massive, slow moving, and ripe for disruption. We are moving fast and optimising for impact.

If you want to build real AI products that people depend on, grow faster than you would anywhere else, and help define a new category, we would love to hear from you.