Who runs Solven Labs
Solven Labs is run by Marcin Walczak, a software engineer and co-founder of ALLDOQ, a UK SaaS platform in the medico-legal space used by HM Courts and Tribunals, instructing solicitors, and around 180 medical experts. ALLDOQ is double-encrypted, OWASP-aligned, and held to the standard a court bundle requires. The same engineering discipline that makes a platform safe enough for a courtroom is what an AI agent needs before it is allowed near production data. That is the work here.
What Solven Labs builds
The work is custom agentic systems with planned tool use, real integrations into the stack a business already runs, and evaluation pipelines that catch regressions before they ship. Not an n8n canvas dressed up as transformation, and not a prompt pasted into a document and called a system. The output is software you own, deployed where your data already lives, with the traces, costs, and failure modes wired to something you can read on a Monday morning. If you want the longer version of how the category works, there is a plain guide to what an AI automation agency actually does.
How a project is run
Every engagement starts by sitting with the people who do the job and watching the boring parts, because the agent's job description comes from the real workflow rather than a wishlist. From there the work is a thin end-to-end slice on real inputs, an eval set built from cases the team has already handled, integration into the existing stack, and a handover with observability so the system is yours to run. The first slice usually takes six to twelve weeks, longer when the integration surface is wide. The reasoning behind scoping it this way is in how to scope an AI automation project, and the honest pricing is in how much an AI automation project costs in the UK.
Where it is based
Solven Labs Ltd is registered in England and Wales and based in Manchester, working with businesses across the UK, including Manchester, Warrington, and Bristol. Day-to-day work is UK hours and async-friendly. You can see the kind of systems this produces in the case studies, or weigh up your own situation with the free AI Readiness Assessment.