Mistral AI buys SkipLabs to make AI-written code more reliable

Representative image: StockSnap / Pixabay
Mistral AI said on 5 November 2026 it is buying SkipLabs, a Paris-London start-up for reliable AI-generated code, its fifth acquisition this year.
What Happened
French AI company Mistral AI said on Thursday, 5 November 2026 that it is acquiring SkipLabs, a small start-up with offices in Paris and London that builds tools to make computer code written by AI more reliable and secure. The price was not disclosed. The SkipLabs team, led by founder Julien Verlaguet, a former Meta engineer, will form a new "Language" team inside Mistral. It is Mistral's fifth acquisition of 2026, according to reports.
Key Facts
- Buyer: Mistral AI, the Paris-based developer of large language models.
- Target: SkipLabs, which works on tools for faster, more reliable and more secure AI-generated code.
- Announcement date: 5 November 2026; financial terms were not disclosed.
- The team, put at about six people in reports, will join the organisation of Guillaume Lample, Mistral's co-founder and chief science officer.
- Its work covers compilers, type systems, static analysis and incremental computation, according to reports.
- Verlaguet said he wants to carry this work into the training of Mistral's models.
- Mistral's earlier 2026 deals were Koyeb, industrial AI firm Emmi AI, advertising specialist Pimento and interface tools maker New Material, reports said.
- In September Mistral announced a €3 billion fundraise that valued it at €21 billion, according to reports.
Why It Matters
AI tools that write code are now widely used by programmers. They save time, but the code they produce can contain mistakes or security holes that are hard to spot. SkipLabs works on the problem from the side of programming languages and their checking tools. A compiler turns human-written code into instructions a computer can run. A type system is a set of rules that catches certain errors, such as treating text as a number, before the program runs. Static analysis means checking code for problems without running it. Incremental computation means recalculating only the parts of a result that changed, which makes tools faster.
By bringing these skills into model training, Mistral is betting that its AI can learn to write code that is correct and safe from the start, rather than relying only on checks afterwards. That matters to large business users: reports on Mistral's recent product updates name the bank BNP Paribas and the shipping group CMA CGM among its showcase customers. The company's own claims about what the team will achieve remain its claims; no product roadmap has been announced.
The deal also shows how Mistral is growing. Rather than only training bigger models, it has been buying small specialist teams through 2026, including an industrial AI firm, an advertising specialist, an interface tools maker and now a programming-language team.
For Indian developers and IT services firms, which write and maintain software for clients worldwide, the deal is a sign that the reliability of AI-written code is becoming a field of competition in its own right.
Peepals calculation: Mistral's €3 billion raise in September at a €21 billion valuation means new investors put in an amount equal to about 14% of the company's value (3 ÷ 21 ≈ 0.143). With SkipLabs as its fifth purchase of 2026, Mistral has averaged roughly one acquisition every two months this year (10 months ÷ 5 deals = 2 months a deal). Our calculation from the reported figures.
| Mistral's 2026 acquisitions | What the target does |
|---|---|
| Koyeb | Not described in reports |
| Emmi AI | Industrial AI |
| Pimento | Advertising |
| New Material | AI interface tools |
| SkipLabs | Reliable, secure AI-generated code |
Impact
Short-term: SkipLabs' small team moves into Mistral's science organisation, and Mistral gains specialists in programming languages and code checking.
Long-term: If the team's methods improve how Mistral's models write code, the company could compete more strongly in AI coding tools.
Who is affected: Mistral and its business customers; software developers who use AI coding assistants; rival AI firms building coding tools; and European technology policy-makers who want home-grown AI companies.
Key Takeaway
Mistral AI's purchase of SkipLabs, announced on 5 November 2026, brings a programming-language team into the company to make AI-written code more reliable, its fifth acquisition this year.
Questions and Answers
What is SkipLabs, which Mistral AI is buying?
SkipLabs is a start-up with offices in Paris and London that builds tools to make AI-generated code faster, more reliable and more secure. It was founded by Julien Verlaguet, a former Meta engineer.
How much did Mistral pay for SkipLabs?
The price was not disclosed when the deal was announced on 5 November 2026.
What will the SkipLabs team do at Mistral?
It will form a new "Language" team in the organisation of co-founder and chief science officer Guillaume Lample, working on compilers, type systems and code analysis, and bringing that work into the training of Mistral's models.
How many companies has Mistral bought in 2026?
SkipLabs is its fifth acquisition of 2026, after Koyeb, Emmi AI, Pimento and New Material, according to reports.
Disclaimer: Prepared by the Peepals newsroom from publicly available sources with AI assistance. Information is accurate to the best of our knowledge at the time of publication and may change. Images may be representative. Not professional advice. Report an error via our contact page.
By Anika Sharma, AI & Innovations EditorCovers AI & Innovation and more for Peepals Tech. All stories by Anika Sharma
Heidi Health layoffs: AI scribe maker cuts 18% as co-founders exit
Nasscom chief Rajesh Nambiar: AI squeezing fresh graduates out of IT jobs
Get the day's stories in your inbox
Peepals Weekly: the week decoded, free. One email a week, unsubscribe in one click.
Anika Sharma is a pen name of the Peepals Global Editorial Team, which researches, fact-checks and publishes every story. The portrait is an illustration.









