Data Scientist

Location: Hybrid. Primarily remote, with in-person collaboration once per month in Manchester or Liverpool

Fletchers Group have a fantastic opportunity to join our growing Team.

Overview

We are looking for a Data Scientist to join our expanding AI team at Fletchers Solicitors. This is a role for someone with a strong analytical and statistical foundation who wants to apply that rigour to building AI systems that operate on complex real‑world legal and medical data.

This role sits within a fast‑paced, high‑autonomy environment where we are rapidly prototyping, iterating, and deploying state‑of‑the‑art AI products for use across Fletchers and the wider legal sector. The team operates as a start‑up‑style unit within the wider business, reporting directly to the executive team, with a clear mandate to move quickly from idea to production.

You will be responsible for understanding our data domain in depth, working closely with legal and clinical experts to formalise specialist knowledge into measurable system behaviour, and designing the evaluation, experimentation, and modelling approaches that underpin our agentic LLM systems. Our products process complex legal and medical documents, surface insights for legal professionals, and increasingly operate autonomously at scale; your work ensures they do so reliably, measurably, and with the analytical rigour that high‑stakes legal output demands. We have ambitious plans to grow the AI team significantly over the next few years as we expand our internal capabilities and commercialise our technology externally.

About Fletchers & the AI Data Science Team

Fletchers is a rapidly growing personal injury and clinical negligence law firm that through recent acquisitions broadened its scope to serious injury and family law. Fletchers was one of the first UK law firms to deploy a machine learning tool supporting legal work (a decision support system for screening incoming cases, launched in 2017) and have a dedicated AI department, which started in August 2023. Since then, we have been consistently widening the range of AI/ML technologies we use at Fletchers, improving existing and productionising new ones through our growing data science & engineering teams.

We believe that we are well positioned to leverage Fletchers' unique and extensive legal‑medical data and industry expertise to build AI services that will shape the future of clinical negligence and personal injury. We are well funded through our Private Equity backers, which allows us to move quickly in the development of sector‑specific AI tools that assist, automate, and augment our specialist legal work. Tools we build range from simple behind‑the‑scene custom ML models to modern user‑facing agentic‑AI UIs. We're also commercialising these tools so that other specialist law firms can leverage Fletchers' legal and AI expertise, and have an ambitious agenda for external growth for the next 18 months.

The Role

As a Data Scientist within the AI team, you will own the analytical and statistical foundation of our LLM‑powered features and agentic pipelines. You will move fluidly between exploring our data domains, scoping new capabilities with the product team and legal stakeholders, eliciting and modelling specialist domain knowledge, designing rigorous evaluation frameworks, and partnering with engineers to ship robust systems into production.

This is a role suited to someone who is technically rigorous, statistically literate, and genuinely curious about both the data and the domain. You will work closely with backend engineers, product stakeholders, and frontend developers across multiple projects, but your distinctive contribution is analytical depth: knowing what to measure, how to measure it, how to interpret it, and what it implies for product and system design.

We are open to candidates ranging from mid to senior level. The scope of responsibilities can flex depending on experience and interests, with clear opportunities for rapid progression and significant professional growth for motivated candidates as the team and product portfolio grow.

Main Responsibilities

• Explore and characterise our legal and medical datasets in depth: profiling distributions, identifying quality issues, and surfacing the patterns that should inform model and pipeline design
• Partner with legal and clinical domain experts to elicit, structure, and formalise specialist knowledge into measurable system behaviour, evaluation criteria, and ground truth datasets
• Design and own rigorous evaluation frameworks for our LLM‑based pipelines: metric design, dataset construction, statistical interpretation of results, and detection of regressions and bias
• Apply statistical and ML methods to analyse model behaviour, quantify uncertainty, run experiments, and inform system design decisions
• Contribute to the design and iteration of agentic LLM pipelines that process complex legal and medical documents, with particular ownership of the modelling, measurement, and analysis layers
• Translate analytical findings into clear, actionable recommendations for product, engineering, and executive stakeholders
• Contribute to the team's shared Python codebase with high‑quality, maintainable code for analysis, evaluation harnesses, and prototype pipelines, taking part in code review and helping establish best practices
• Stay close to the fast‑moving AI and applied‑ML literature, identifying methods and techniques with genuine relevance to our domain and bringing them back into the team

You may be a good fit if you have:

• A strong quantitative background, typically a postgraduate degree (MSc or PhD) in a numerate discipline such as statistics, machine learning, computer science, mathematics, physics, computational linguistics, or a closely related field, or equivalent applied research experience
• Strong proficiency in Python and the standard scientific stack for data exploration, modelling, and analysis
• Solid grounding in statistics and applied ML: hypothesis testing, experimental design, evaluation methodology, and reasoning about uncertainty and small samples
• Hands‑on experience working with LLMs in an applied setting: prompting, evaluation, RAG, structured outputs, or agent design. You don't need to have shipped to production at scale, but you should have built and measured something real
• A delivery‑oriented mindset: you take ownership of outcomes, are happy to write production‑quality code where needed, and can drive work from an ambiguous problem statement to a validated result with minimal supervision
• Comfort working with non‑technical stakeholders, eliciting domain knowledge, and translating between business questions and quantitative methods
• Based in the UK, with the ability to attend in‑person team meetings approximately once per month
• Fluency in English

Strong candidates may also have:

• A PhD in a relevant discipline, or equivalent experience leading applied research or analytical projects
• Familiarity with our core stack: PydanticAI, Azure OpenAI, and Azure infrastructure (Container Apps, Azure AI Foundry, Azure Document Intelligence)
• 3+ years of experience in a data science, applied ML, or applied research role
• Experience designing evaluation frameworks for LLM‑based or other ML systems, including labelling strategies, inter‑annotator agreement, and reconciliation of offline and online metrics
• Exposure to document processing pipelines: OCR, document parsing, structured information extraction from unstructured legal or medical text
• Familiarity with software development tooling supporting day‑to‑day delivery: version control, CI/CD pipelines, testing, code review, pre‑commit hooks
• Start‑up or greenfield experience, particularly building AI features or data pipelines from scratch in a fast‑moving environment
• Previous experience in a regulated or high‑stakes domain such as legal, medical, or financial services, where output quality and auditability matter

We encourage you to apply even if you do not believe you meet every single qualification. We mean it when we say strong candidates don't need to tick every box. Job descriptions are wish lists, not checklists. If the work excites you and you have most of what we're looking for, we'd rather see your application than have you talk yourself out of it.

We want to grow quickly, so reserve the right to close this role early depending on candidate quality and volume.

In return Fletchers offers:

• 35 hours working week
• Bonus scheme (subject to targets being met)
• 35 days holiday a year including bank holidays and Christmas Closure, you also have an option to purchase up to 3 extra days’ holiday per year, until you long service award the 3 days after your 5th, 7th, and 9th full year of employment
• Pension scheme with tax‑efficient salary sacrifice option
• Life Assurance Policy
• Medicash policy to cover some dental, optical and other medical expenses
• Company sick pay
• Enhanced Maternity, Paternity and IVF schemes
• Flexible, agile working environment with a positive work‑life balance
• Varied calendar of funded company social events

Fletchers Group is an inclusive employer with a diverse workforce: we welcome applications from any and all backgrounds. We’re committed to providing a culture and environment where everyone can thrive, understanding and celebrating that no colleague is the same. We’re proud of our work on the Disability Confident scheme, signed up to BITC's Race at Work charter, and working towards Henpicked's Menopause and Menstrual Friendly accreditations.

#J-18808-Ljbffr