Fractional General Counsel, Norven

Norven | London, United Kingdom (Hybrid) | Contract | Approx. 2 days per week | Start October 2026

Proposed engagement: approximately two days per week for an initial eight-week paid engagement, with scope to continue as the work develops. UK-based preferred, remote working with regular UK-hours collaboration. Fee and scope agreed with the successful candidate.

ABOUT NORVEN

Norven is an early-stage UK company building the evidence infrastructure that lets organisations deploy autonomous AI agents with confidence. Governments, institutions and corporations increasingly want AI systems to take real actions, but cannot show what those systems are permitted to do, that the limits hold in the real workflow, or that the evidence would stand up to a board, an auditor, an insurer or a regulator. We are building the product that answers those questions. Our first paid work is in UK financial services; the ambition is a system of record for consequential AI across sectors and jurisdictions, including the EU, the US and the GCC, where the obligations, the regulators and the pace of change are all different.

We are two founders. (CEO) has nineteen years in executive search and recruiting scarce senior expertise. (CTO) is a physicist who has spent a decade putting AI models into production in healthcare and drug discovery. We are pre-revenue, with a working prototype, and preparing for our first paid design partners.

THE ROLE This is not a General Counsel role and it is not a contracts role. It is the legal expertise that goes into the product itself.

Our product is only as good as the legal judgement embedded in it: which obligations a given AI action engages, what facts are needed to decide, and where a conclusion cannot be automated and a person must decide. An organisation that lets an AI agent act is exposed to the whole body of law that governs its own conduct: data protection and privacy, sector regulation, AI-specific legislation, financial-crime and payments rules, consumer and equality law, contractual and professional duties, and the differences between all of these across jurisdictions. No lawyer can assess every case, so the job is to decide what can be templated, build the framework for capturing and structuring obligations so it can grow across sectors and jurisdictions, write the first rules, and draw the line where human judgement takes over. You will work directly with the CTO from the prototype stage.

WHAT YOU WILL DO Design the framework by which legal and regulatory obligations are identified, structured and mapped to what an AI system is permitted to do, so that it can extend from the first use case to new sectors, workflows and jurisdictions. Define the legal scope for an initial, tightly bounded use case in UK financial services, and translate the relevant obligations (for example UK GDPR, FCA conduct and systems rules, payments and financial-crime requirements, the EU AI Act) into structured decision rules, control requirements and test scenarios with the technical team. Document assumptions, exceptions, source material and the limits of what an automated assessment can establish, so the product's outputs and our customer claims never go further than the evidence. Work with the founders to set risk appetite, and design the human review and escalation criteria that follow from it. Help validate product outputs against representative cases and build a maintainable approach to updating legal content as legislation, regulatory guidance and facts change, including the wave of AI regulation now arriving across the UK, EU and US. Set the standard for the expert reviewers who follow you, and help us identify the specialist counsel and regulatory advisers we will need as we move into new sectors and jurisdictions. Review the demo and any customer-facing description of capability before it is shown, with our legal and commercial advisers.

WHO WE ARE LOOKING FOR One trait runs through everyone we want to talk to: you have already turned law into something a system applies, and you know what it got wrong. Three backgrounds fit particularly well. The legal engineer. You work or have worked in a regtech or legaltech company as a Legal Engineer, Legal Knowledge Engineer, Head of Legal Engineering or Product Counsel. You have encoded obligations into a rules-based or AI product and think naturally in templates, edge cases and referral-to-human design. The AI governance counsel. You are in-house at a scale-up or enterprise deploying AI at scale, or you moved from privacy counsel into AI governance. You have run AI risk assessments, built model-governance frameworks and sat with engineering on real deployments. You know the EU AI Act, ICO guidance and the UK sector-regulator approach in practice. The regulated-sector technologist. You come from the legal or compliance function of a bank, insurer, asset manager, healthcare provider or other regulated business where you built automated controls, surveillance or authorisation logic, or you are regulator, government or Big Four alumni who moved into a product role. You speak the language of accountable owners, release committees and evidence.

If you are a privacy lawyer who has done DPIAs on automated decision-making and want to go deeper into agentic systems, we would like to hear from you too.

WHAT YOU BRING A qualified lawyer (England and Wales preferred) with 8+ years' PQE, mostly in-house or in a regtech, legaltech or compliance-technology business. Work you personally delivered that turned legal or regulatory requirements into a functioning workflow, product feature, rule set or control, and can explain. Breadth across the law that governs organisational conduct, and the judgement to know which parts an autonomous system engages. Deep expertise in at least one of: data protection and privacy, financial-services or another sector's regulation, or AI-specific regulation, and the curiosity to build out the rest with specialist support. Working knowledge of the EU AI Act, the UK's approach to AI regulation and the direction of travel internationally. Strong technical literacy: comfortable with data flows, access controls and cloud products, and with arguing about edge cases in a design session with engineers. Comfort with bounded, imperfect systems. You can question assumptions, make uncertainty explicit, and design the caveats and referral rules that make an automated legal judgement honest. Availability of around two days a week from October, including part of Monday to Wednesday to collaborate with the CTO, UK time zone, occasionally in London.

The initial engagement is practical and bounded to one workflow. You will not be expected to be the expert in every jurisdiction or sector yourself, or to build the underlying software; you will be expected to design an approach that can absorb them.

WHAT THE INITIAL ENGAGEMENT SHOULD ACHIEVE A framework for identifying and structuring obligations that can extend across sectors and jurisdictions; an agreed legal scope for one workflow; an initial library of reviewed rules and test cases; clear human-review boundaries; and a documented approach to validating, maintaining and explaining the resulting product outputs.

WHAT WE OFFER A paid founding-stage fractional engagement (day rate or project fee), reviewed at the end of the initial period, with scope for deeper involvement and upside as the company reaches its next commercial and funding milestones. Direct work with the CTO on a problem no one has templated yet, in a company whose product depends on your judgement. A small, candid founding team that would rather hear "that cannot be expressed as a rule" than a confident guess.

HOW TO APPLY Apply through LinkedIn or email enquiries@norven.ai the subject "Product Counsel / Legal Engineer". Please include your CV or LinkedIn profile, two short examples of relevant work (sanitised; no confidential client material), your professional qualification and jurisdiction, your availability, and your proposed day rate or project fee. No cover letter needed.

We will review relevant experience before arranging a 25-minute first call with the CEO, followed by a 45-minute second call with the CEO and CTO. Any substantive practical assignment will be agreed and paid.