Custom AI Infrastructure

Jan 1, 2025 · 1 min read
projects

Custom AI Infrastructure project involving low-level optimization of Deep Learning frameworks.

Details

  • Objective: Enable support for next-gen NVIDIA hardware (RTX 5070 Ti / Blackwell Architecture) pending official upstream support.
  • Work: Built custom TensorFlow wheels from source with specific CUDA toolkit versions and C++ compiler flags.
  • Stack: C++, CUDA, Linux, Bazel/Python Build Systems.
CUDA C++ Linux Python Build Systems
Authors
Lakshya (he/him)
AI Researcher & Systems Engineer
M.Tech AI & ML student specializing in Multimodal Representation Learning and Advanced Systems Engineering.