
Academic lab · Zurich, Switzerland
SRI LabETH Zurich
ETH Zurich's group on trustworthy machine learning, producing neural network verifiers, LMQL and ML-for-code tools.
Zurich, Switzerland
- Type
- Academic lab
- Based in
- Zurich, Switzerland
- Key papers
- 1
- Featured projects
- 4
About
The Secure, Reliable and Intelligent Systems Lab at ETH Zurich is led by Martin Vechev. It works on program analysis, verification and the reliability of machine learning.
Its AI2 paper at IEEE S&P 2018 introduced abstract interpretation for certifying neural networks, and the ETH Robustness Analyzer (ERAN) followed.
Its code is published on GitHub, including LMQL, a language for programming with language models.
Mission and ambitionsThe lab works to make intelligent systems safe, secure and reliable by combining machine learning with program analysis.
Known for
- 01
AI2 neural network certification
- 02
ERAN robustness analyzer
- 03
LMQL query language
Milestones
- 2018
AI2
Safety and robustness certification with abstract interpretation at IEEE S&P.
- 2019
ERAN released
Robustness analyzer for deep neural networks.
- 2020
DeepCode and ChainSecurity acquired
Vechev's spinouts DeepCode (by Snyk) and ChainSecurity (by PwC) were acquired; LatticeFlow was co-founded the same year.
- 2023
LMQL
Language-model programming language published.
Open source
A language for constraint-guided and efficient LLM programming. (4,228 stars)
Securify v2.0 (636 stars)
Machine Learning to Deobfuscate Binaries (453 stars)
ETH Robustness Analyzer for Deep Neural Networks (348 stars)
Research papers
Notable releases
Openness
Tools are public on GitHub, mostly under open-source licences.
People
- Martin VechevProfessor and head of the lab
Founder of DeepCode, ChainSecurity and LatticeFlow; architect of INSAIT.
- Timon GehrSenior scientist
Senior scientist at the lab.
Listed from the lab's own pages and public profiles.
Contact and community
Contact via site.
Sources
Compiled from the lab's public pages, Hugging Face and GitHub. Last checked 2026-11-17. Something wrong or out of date? Tell us on Discord.



