Platon I. Karpov
New York, New York, United States
Summary
👤
SeniorPlaton Karpov is a Sr. Machine Learning Engineer in New York with nine years of experience bridging astrophysics, high-performance computing, and production ML for finance and healthcare. Trained as an astrophysicist with a PhD from UC Santa Cruz, he pioneered physics-informed ML for turbulence in core-collapse supernovae, open-sourcing an ML-to-Fortran integration pipeline (Sapsan) used to embed PyTorch models in legacy hydrodynamics codes. At Provectus he leads teams productionalizing GenAI, specializing in LLM research, tuning, continuous evaluation, RAG systems and hybrid knowledge-graph integrations. Equally comfortable on GPUs and supercomputers, he combines deep scientific modeling experience with pragmatic engineering to turn complex physics and regulatory use-cases into reliable, auditable ML deployments.
10 years of coding experience
Github Skills (15)
machine-learning8
pytorch8
application8
fine-tuning8
astrophysics8
llm7
ocr7
vector-database7
llama-index6
3d5
keras4
visualization3
tensorboard3
tensorflow2
neural-network2
Programming languages (2)
Github contributions (5)
pikarpov/sapsan-web
Jul 2020 - Aug 2022
Contributions:9 commits, 2 pushes in 2 years 1 month
pikarpov-LANL/COLLAPSO1D
Jul 2020 - Mar 2023
Embedding 3D Turbulence into 1D Core-Collapse Supernovae with ML
Contributions:191 commits, 10 PRs, 5 pushes in 2 years 8 months
3dmachine-learningsupernovae
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.