AI Engineer - Wexler

Praxis, Inc.; Greater London; On-site; GBP 70,000 - 90,000; Full time

Benefits offered by this job

Competitive salary & equity

Autonomy & ownership

Learning & development budget

Bi-annual team retreats

Job summary

Wexler is building the leading gen-AI platform for litigation, headquartered in London. We seek an exceptional AI engineer to shape the future of legal work and oversee the AI strategy, crafting agentic systems that analyse millions of documents to extract critical factual information for winning cases.

You will design LLM agents, develop robust evaluation pipelines, and drive prompt engineering while balancing token limits, cost and latency.

Qualifications

• Expert-level Python with a track record of shipping production LLM systems.
• Hands-on experience designing LLM agents and RAG pipelines.
• Proficiency with AI evaluation frameworks and end-to-end evaluation pipelines.
• Solid grounding in NLP fundamentals (tokenisation, embeddings, bias handling).

Responsibilities

• Design, build and refine production-grade LLM agents and proprietary algorithms for expert litigators.
• Lead development of robust evaluation frameworks and pipelines, establishing metrics and benchmarking models on large datasets to steer R&D.
• Drive advanced prompt-engineering practice to maximise model efficacy.
• Apply core NLP fundamentals (tokenisation, embeddings, model architecture) while keeping outputs structured, performant and cost-efficient.
• Own clean, modular Python back-end code for data-intensive systems, balancing real-world constraints such as token limits, cost and latency.

Skills

Python

LLM development

NLP fundamentals

Prompt engineering

AI evaluation

Tools

RAG pipelines

Production back-end

Job description

About Wexler

Wexler is building the best AI system for litigation on the planet. We work with some of the world’s largest law firms, helping them to solve their most complex cases, find the winning strategy in each matter, and create clarity from the chaos of documents and facts in each case. We are a rapidly growing, legal AI company based in London. Clifford Chance, HSFKramer, Goodwin, Addleshaw Goddard and more rely on us to help find the critical facts that can determine their chances of winning a case.

We’re backed with a $5.3M Seed by Pear VC, Seedcamp, The LegalTechFund + many leading industry angels.

We are building a comprehensive AI platform for managing, resolving and preventing legal disputes across Enterprise law firms and Fortune500 companies. We are growing 10x YoY and signing more eminent firms every month. Our system extracts, objective, cross referenced facts from millions of documents, helping litigators win more cases whilst saving months of working hours. In this way we ensure every client gets the representation they deserve.

AI is transforming the law, but most tools have focused on contract law or are generalist copilots that aggregate the tasks lawyers do across practices. Wexler is the leading gen-AI platform specifically built for the nuances of litigation, and our growth proves the story is resonating.

About the Role

We are looking for an exceptional AI engineer to shape the future of legal work. We are building agentic systems that analyse millions of documents to extract the critical factual information that wins legal battles. As a core member of the team, you will have direct oversight of the AI strategy, manipulating the most powerful technology in software to solve real problems faced by some of the world’s most expert litigators.

Responsibilities

• Design, build and refine production-grade LLM agents and proprietary algorithms for expert litigators
• Lead development of robust evaluation frameworks and pipelines, establishing metrics and benchmarking models on large datasets to steer R&D.
• Drive advanced prompt-engineering practice to maximise model efficacy.
• Apply core NLP fundamentals (tokenisation, embeddings, model architecture) while keeping outputs structured, performant and cost-efficient.
• Own clean, modular Python back-end code for data-intensive systems, balancing real-world constraints such as token limits, cost and latency.

Technical Must-Haves

• Expert-level Python with a track record of shipping production LLM systems
• Hands-on experience designing LLM agents and RAG pipelines.
• Proficiency with AI evaluation frameworks and end-to-end evaluation pipelines.
• Solid grounding in NLP fundamentals (tokenisation, embeddings, bias handling).
• Ability to juggle token limits, cost and latency while delivering structured outputs.

You will be:

A builder. You might spend your weekends at hackathons or playing around with DeepSeek. You love working with LLMs to crystallise new ideas and make them reality

Excited to work in our office in London. We work from the WeWork in Aldwych, and we spend 4 days a week in person.

Excited to work with a group of kind, ambitious builders solving challenging problems for enterprise clients.

Fascinated with AI and LLMs. You want to understand how they are made, what you can do with them, and where they are going. You will come to the office with new ideas which we can turn into production in a matter of hours.

Ready to work in a fast-paced startup environment with evolving requirements and priorities

What We Offer

• Competitive salary and meaningful early-stage equity (£70-90K)
• Huge autonomy and ownership. You will be designing and realising AI systems for some of the most expert users in AI.
• Budget for learning and professional growth
• Bi-annual team retreats

About the opportunity:

Legal AI has caught the attention of VCs and the wider tech world, but the reality is adoption with users is low in the enterprise, with roughly 20% of lawyers using AI tech. Of those lawyers, most are transactional or contract lawyers and they use generalist platforms that look to aggregate the tasks lawyers do but do not go deep into specific workflows. That’s where Wexler is different. We are a ‘scalpel’ rather than a ‘Swiss Army knife’, meaning we go deep into the workflows that can decide how a legal case is won. We have extremely positive feedback from lawyers, and we have an ambitious roadmap of features to expand our coverage while retaining the forensic detail that makes our product great.