Dhruv Jain

Lead ML Engineer

Santa Clara, California, United States10 yrs experience

Highly Stable

Key Highlights

Expert in compiler infrastructure for AI hardware.
Led significant ML and hardware integration projects.
Proficient in large-scale C/C++ codebases.

Stackforce AI infers this person is a Compiler Engineer specializing in AI hardware and machine learning systems.

Contact

dhruvjain.1027@gmail.com LinkedIn

Skills

Core Skills

CompilersC++Machine Learning

Other Skills

EngineeringHaskellLLVMMLIRPyTorchPythonTensorFlow

About

Building state-of-the-art compiler infrastructure for AI hardware accelerator (for training & inference). Specialties: Huge C/C++ codebases, LLVM's MLIR (Multi-Level Intermediate Representation), ML & data science in using Python (Pandas, NumPy), ML frameworks (PyTorch, TensorFlow), debugging & profiling tools. Areas: compilers, network protocol stacks (QUIC, UDP), linux kernel development, machine-learning system development, data science.

Experience

Sambanova systems

Senior Compiler Engineer

Sep 2022 – Present · 3 yrs 6 mos · Palo Alto, California, United States

Developing an embedded DSL in C++ using MLIR built for implementing custom ML kernels on proprietary CGRA (Coarse-Grained Reconfigurable Architecture) chip.Developing an embedded DSL in C++ using MLIR built for implementing custom ML kernels on proprietary CGRA (Coarse-Grained Reconfigurable Architecture) chip.
Skills: LLVM · C++ · MLIR · Compilers

LLVMC++MLIRCompilers

Google

Senior Compiler Engineer

Mar 2016 – Sep 2022 · 6 yrs 6 mos · Sunnyvale, California, United States

Google Brain: Technical Lead in a stealth mode research project at the intersection of Machine Learning and hardware (Haskell, MLIR, a little bit of Bluespec).
Google Core ML Compilers: I worked on MHLO, the shared compiler IR for XLA-Next, IREE, EdgeTPU and others. My primary workstream is bringing it and the HLO IR from XLA-Legacy into semantic equivalence. I also worked on Tensorflow on GPU (C++, Python, MLIR).Google Brain: Technical Lead in a stealth mode research project at the intersection of Machine Learning and hardware (Haskell, MLIR, a little bit of Bluespec). Google Core ML Compilers: I worked on MHLO, the shared compiler IR for XLA-Next, IREE, EdgeTPU and others. My primary workstream is bringing it and the HLO IR from XLA-Legacy into semantic equivalence. I also worked on Tensorflow on GPU (C++, Python, MLIR).
Skills: MLIR · Haskell · Python · Compilers

MLIRHaskellPythonCompilersMachine Learning