Ahmed Irfan is a Principal Research Scientist based in the San Francisco Bay Area with 12 years of experience advancing formal verification, SMT solving, model checking and neuro-symbolic AI across academia and industry. He has held research and applied roles at Stanford, SRI, AWS and a stealth AI startup, and now leads rigorous, scalable verification research at Code Metal. Ahmed contributes to notable open-source tooling—improving pysmt’s rewriting and testing infrastructure and enhancing Yosys’s BTOR backend—demonstrating a knack for turning deep theory into practical tooling. His background combines a PhD in ICT and a postdoc in computer science with hands-on systems engineering experience, enabling him to bridge proofs, solvers, and production constraints. Colleagues rely on him for precise, reproducible solutions to hard automated-reasoning problems and for shipping backend innovations that improve solver robustness.
12 years of coding experience
16 years of employment as a software developer
Doctor of Philosophy - PhD, Information and Communication Technology, Doctor of Philosophy - PhD, Information and Communication Technology at Università di Trento
Master of Science - MS, Computational Logic, Master of Science - MS, Computational Logic at Technische Universität Dresden
Bachelor of Science - BS, Computer Science, Bachelor of Science - BS, Computer Science at National University of Computer and Emerging Sciences
Postdoc, Computer Science, Postdoc, Computer Science at Stanford University
pySMT: A library for SMT formulae manipulation and solving
Role in this project:
Backend Developer & Test Automation Engineer
Contributions:18 commits, 2 PRs, 18 comments in 3 months
Contributions summary:Ahmed significantly contributed to the pysmt library, focusing on its core functionalities and testing infrastructure. Their work included the addition of a "toplevel-propagation" function within the rewriting module, which aims to simplify and optimize SMT formulas. They also wrote multiple tests for this new function, ensuring its correctness and validating its behavior against example formulas. Furthermore, the user refactored simplification routines, specifically for the handling of multiplication in order to ensure the fix point is reached in more cases.
Contributions summary:Ahmed primarily focused on enhancing the BTOR backend for the Yosys Open SYnthesis Suite. Their commits introduced new features such as the -driver feature and added support for the $pmux cell translation, expanding the tool's capabilities. They also addressed and corrected several bugs related to memory handling, XNOR, and logic_not translations, leading to improved accuracy and reliability of the backend. These modifications indicate a focus on improving the BTOR backend's functionality and compatibility with the Yosys suite.
yosys
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.