Snapshard is an expert ML engineering practice. Production systems at scale.

Ilyas leads every Snapshard engagement end-to-end. The practice exists because the hardest parts of shipping ML (making research reproducible at scale, getting inference fast and cheap enough to deploy, building infrastructure that survives a year of iteration) are still the parts most teams underbudget.

MSc in Statistical Science from the University of Oxford and an engineering degree from École Polytechnique (Major in Applied Mathematics). Research published at NeurIPS 2022 (distributional reinforcement learning, with IBM Research) and ICML 2021 (amortized Bayesian experimental design, with the Oxford CS group).

Prior work: AI Research at IBM Research Singapore, Applied Scientist on Amazon's Supply Chain Science team, and three years as AI R&D Lead at Arcturus Studio (3D deep learning, graph neural networks, texture super-resolution). Concurrent venture: co-founder and lead ML engineer at Gradients, a distributed AutoML platform for LLM and diffusion fine-tuning.