Artificial intelligence has changed the way we design and ship products. It is now part of the daily work of every discipline - engineering, design, product - and of our own operations. Here is exactly what we do with it, which tools we use, and what it changes for your projects.
We share our in-house expertise through training sessions, complemented by talks from outside experts.
Privileged access to one of the world's leading AI labs, so we can integrate the best reasoning models.
Every week, a dedicated team meeting is given over to sharing and analysing the latest advances in AI.
Our software and product engineers orchestrate agents that work in parallel. They run AI-driven quality control loops covering security, scalability and testing. Typical uses: speeding up development, extracting domain knowledge from undocumented legacy code, diagnosing and fixing bugs, reviewing pull requests, implementing designs and checking the result visually.
Our designers work from a stack built on a design system and a prompt of business rules specific to each client. With that context, their tools generate several visual directions per feature and adapt an interface across devices in minutes. AI also produces realistic, lopsided or extreme data to stress-test a layout. From ideation to high-fidelity mockups, by way of visuals created with the latest image generation models, AI runs through the whole chain.
Our product managers use Deep Research to scan the web and benchmark competing solutions. They build specifications that can be queried in plain language with NotebookLM. Within hours they produce first interface drafts and testable prototypes with Figma Make and Lovable. They also drive tracking tools (roadmap, Kanban, meeting notes) and analyse product usage metrics to advise their clients.
InvoicingBalin is an autonomous agent that talks to the team, understands the client context (contracts, invoices, specific rules) and takes action: issuing invoices and credit notes, managing the schedule and sending emails. It handles several million euros of invoicing every month.
MCPThe company's internal data is available over MCP (Model Context Protocol) to everyone on the team, with proper access control, so that knowledge can be put to the best possible use.
RecruitmentThorin sources relevant candidates from a plain-language brief and schedules interviews on its own.
We work with every kind of client, across every industry. Whether you are an entrepreneur or lead a large organisation, we put together a team that fits your need.
More than 800 companies have trusted us to build their web, mobile and AI products
Your request
We'll get back to you within the hour.
