Jul 2022 – Jan 2023 · 6 mos · Pune, Maharashtra, India

• Develop nvidia internal CUDA profiler tool.
• Design metrics required for the profiler tool.
• Setup process for validation of GPU counters.

Nov 2017 – Jun 2022 · 4 yrs 7 mos · Pune, Maharashtra, India

• Develop Nvidia internal profiler tool to help deep learning architects optimize their code.
• Architect couple of GPU profiler features.
• Design GPU counters/metrics required for new hardware features.
• Setup process for validation of the GPU counters and metrics.

Dec 2009 – Jun 2017 · 7 yrs 6 mos · Pune, Maharashtra, India

• Worked primarily on CUDA profiler tool nvprof.
• Participated in designing of CUDA profiler APIs (CUPTI - https://docs.nvidia.com/cupti/index.html) and implemented the initial version of library and ported for multiple GPU platforms.
• Designed GPU metrics and various profiler features for multiple GPU architectures.
• Emulation and bringup of various profiler features for multiple GPU architectures.

Sep 2007 – Nov 2009 · 2 yrs 2 mos · Pune, Maharashtra, India

• Section manager of Direct 3D team in Pune.

Sep 2005 – Sep 2007 · 2 yrs · Pune, Maharashtra, India

• Developed direct show filters for various audio and video codecs and helped in integration of all components of media player for nvidia's mobile chip.

Dec 2000 – Aug 2005 · 4 yrs 8 mos · Pune, Maharashtra, India · On-site

• Developed video/imaging codecs in C and port them on TI's C55x DSP chip.
• Developed media framework and ported on TI's OMAP chip (ARM+DSP).

Oct 1999 – Dec 2000 · 1 yr 2 mos · Pune, Maharashtra, India · On-site

• Ported insurance software to use relational database from network database.