Shalva Mist
Sr. Staff FAE - AI IP at Tenstorrent
Redmond, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolShalva Mist is a senior staff field applications engineer specializing in AI IP who blends hands-on embedded software expertise with customer-facing technical leadership across SoC, FPGA, and simulation environments. With a career spanning CEVA, Microsoft, Cadence, and now Tenstorrent, he has taken IP from initial integration through to shipping mass-production silicon and strategic account engagements. He is proficient in embedded DSP firmware, core drivers, C/C++, Python, and SystemC simulations, and has practical experience optimizing performance and memory at assembly level. While at Microsoft he contributed to the high-profile ONNX Runtime—adding JS/C++ fixes, XNNPACK GEMM/MATMUL integrations, and E2E Image Tensor tests—demonstrating cross-language full-stack impact on an industry-standard ML runtime. Based in Redmond, he combines deep product knowledge with customer training and tooling creation to accelerate adoption and debug complex integration challenges. Known for translating low-level optimizations into customer-facing value, he thrives at the intersection of applied R&D and field engineering.
4 years of coding experience
10 years of employment as a software developer
Engineer's Degree Electrical and Electronics Engineering, Engineer's Degree Electrical and Electronics Engineering at Ben-Gurion University of the Negev
English, Hebrew, Georgian
Github Skills (13)
xpack10
javascript10
onnx10
wasm-pack9
rust-wasm9
cprogramming-language9
machine-learning9
webassembly9
c-language9
deep-learning8
deeplearning-ai8
hardware-acceleration8
tensorflow8
Programming languages (4)
Github contributions (5)
microsoft/onnxruntime
Apr 2022 - Jan 2023
ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator
Role in this project:
Full-stack Developer
Full-stack DeveloperContributions:50 reviews, 43 commits, 40 PRs in 9 months
Contributions summary:Shalva contributed to both the JavaScript (JS) and C++ components of the ONNX Runtime project. They addressed permission issues in JavaScript by changing symbolic link creation and updated the C API to handle non-tensor data correctly. Further, the user integrated XNNPACK GEMM & MATMUL ops which improved performance and added scripts for simpler web builds. They also added an E2E test for the Image Tensor API.
runtimetrainingtensorflowai-frameworkaccelerator
shalvamist/OllamaRAG
Dec 2024 - Mar 2025
Streamlit app leveraging Ollama to do local RAG
Contributions:4 PRs, 56 pushes, 4 branches in 3 months
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.