Sayan Dey

AI Researcher

India2 yrs 10 mos experience

AI EnabledAI ML Practitioner

Key Highlights

Senior AI Compiler Engineer with extensive experience in AI and HPC.
Proven track record in optimizing AI models for edge devices.
Strong academic background with MTech from IIT Hyderabad.

Stackforce AI infers this person is a specialist in AI and Compiler Engineering with a focus on Healthcare applications.

Contact

Skills

Core Skills

C++PytorchAi For Healthcare

Other Skills

Multimodal LLMsPython (Programming Language)LLM FinetuningHigh Performance Computing (HPC)Deep LearningKernel ProgrammingLLVMINDUCTORGitLarge Language Models (LLM)GNU/LinuxDockerClangGenerative AICompiler Optimization

Experience

2 yrs 10 mos

Total Experience

10 mos

Average Tenure

2 mos

Current Experience

Amd

Senior SDE

Apr 2026 – Present · 2 mos · Hyderabad, Telangana, India · Hybrid

Qualcomm

Engineer

Feb 2025 – Apr 2026 · 1 yr 2 mos · Bengaluru, Karnataka, India · Hybrid

Enabled Pytorch operators on Qualcomm’s AI 100 accelerator to support open-sourced LLMs.
Developed vectorized kernels using Qualcomm Hexagon Vector eXtensions (HVX) intrinsics.
Optimized scalar kernels with Multi NSP support using custom tiling algorithms

C++PyTorch

Global innovation hub (gih)

Gen AI Research Intern

Jul 2024 – Sep 2024 · 2 mos · Hyderabad, Telangana, India · Remote

Implemented a Knowledge Distillation framework to support model deployment in edge devices. Used MedMCQA dataset with qwen2-1.5B as teacher model and qwen2-0.5B as student model.
Worked on Model Explainability and interpreted the output of Google’s gemma2 model by visualizing the activation features of input tokens at each layer.
Performed Benchmarking on text and video datasets using different LLMs from OpenAI and Google GenAI.
Extracted Time-Series data from ECG images and processed it to get model prediction about the patient.

AI for HealthcareMultimodal LLMs