Machine Learning Systems Engineer at Qualcomm

I am a Machine Learning Systems Engineer at Qualcomm since 2021, currently working on the Qualcomm AI 100 product suite. My work focuses on high-performance LLM inference serving, where I leverage and contribute to frameworks like vLLM and LLM-D to optimize model deployment on specialized AI hardware.

Previously, I was a Software Development Engineer at Ericsson, where I specialized in testing Virtual Network Functions (VNF) and tackling cloud-native challenges within the telecom sector.

My technical interests have evolved towards the cutting edge of Generative AI. I am particularly passionate about:

• Distributed Inference: Scaling LLM serving across multiple accelerators and nodes.
• LLM Fine-tuning: Exploring efficient adaptation techniques (PEFT/LoRA) for large-scale models.
• AI Systems: Optimizing the intersection of hardware and software for deep learning workloads.

I enjoy playing Table Tennis, fixing broken electronics, watching movies, and programming.

---

Publications

Accepted Publications

Tracking Inbound Enemy Missile for Interception from Target Aircraft Using Extended Kalman Filter
Gokkul Nath T.S., Sudheesh P., Jayakumar M
Mueller P., Thampi S., Alam Bhuiyan M., Ko R., Doss R., Alcaraz Calero J. (eds) Security in Computing and Communications. SSCC 2016. Communications in Computer and Information Science, vol 625. Springer, Singapore. Link

Preprints

A Novel Clustered Support Vector Machine with Reduced Support Vectors for Big Data Classification
Gokkulnath T.S., Ramanathan R
Reduction in the number of Support Vectors is essential because it reduces the computational complexity of the model, which in turn gives the user the ability to implement real-time applications on low-power computing devices and reduces hardware requirements. Techrxiv

Up-to-date list of publications can be found at Google Scholar: Link