Akshay P P

About Akshay P P

I am an ML systems engineer working at the intersection of deep learning models and AI accelerator hardware.Currently at Ignitarium, where my primary focus is deploying and validating state-of-the-art deep learning models on Tenstorrent’s custom AI accelerators using the TT-Metal software stack.My work involves:• Porting and deploying complex vision and multimodal models onto Wormhole and Blackhole architectures• Adapting model graphs to accelerator-specific execution constraints• Debugging functional and performance issues across compiler, runtime, and kernel layers• Validating numerical correctness and performance against reference frameworksI have hands-on experience working with transformer-based models, perception networks, and large vision pipelines, and I enjoy solving low-level performance and systems challenges that sit close to the hardware.I’m particularly interested in AI compilers, accelerator runtimes, and scalable deployment of ML models on next-generation hardware.

Experience

1. Senior Engineer

   Ignitarium

   Dec 2024 — Present · Kochi, IN

   Deployed and validated deep learning models on AI accelerators• Worked on model bring-up and execution for Client architectures• Ported vision and transformer-based models by adapting computational graphs to accelerator constraints• Debugged runtime, kernel-level, and numerical issues during model execution• Performed correctness validation and performance analysis against reference PyTorch modelsModels worked on include: GPT-OSS, Panoptic-DeepLab, Transfuser, MapTR, Efficientdet D0, YOLO variants, Stable Diffusion components

Education

• Government Engineering College, Thrissur

  Bachelor of Technology - BTech, Electronics and Communications Engineering

  2015 — 2019

• NATIONAL INSTITUTE OF ELECTRONICS & INFORMATION TECHNOLOGY (NIELIT)

  Master of Technology - MTech, embedded systems

  2020 — 2022