Postdoc in AI-based formally verified code generation

Job description

AI is rapidly becoming the primary author of the world's software. Today's models generate code at a speed no human can match but how can we trust code we did not write ourselves? Formal verification offers a path to mathematically guaranteed correctness, yet combining it with AI-based code generation raises deep and largely unsolved challenges.

As a postdoctoral researcher, you will tackle exactly this question: how to generate code with AI and formally verify that it does what it should. Your work will combine modern AI/LLM-based code generation with formal methods to produce software that is both fast to create and provably trustworthy.

The position is part of SAFIR (Secure AI for Intelligent Resilience and Confidentiality in the Cloud), a EUR 2.5M collaborative project between KTH, Saab, Nvidia, Ericsson, Red Hat, CanaryBit and RISE, building next-generation secure and dependable AI for critical domains such as telecom, defence and...