One-Day Workshop
IISc–Qualcomm Workshop on
AI Acceleration & Inferencing
About the Workshop
AI at the Edge
Explore architectures and frameworks for deploying large AI models efficiently on edge devices with constrained power and compute budgets.
Compilers & Accelerators
Dive into compiler design for AI accelerators and custom hardware platforms for next-generation neural network inference.
LLM Inferencing
Learn about KV-cache optimisation, distributed inference strategies, and scheduling techniques that make LLM serving fast and scalable.
Schedule
| Time | Topic | Speaker |
|---|---|---|
| 08:30 – 09:00 | AI Workloads Past, Present & Future: Impact on Compute Architecture— Anuj Gupta & Nitin Jain | Anuj Gupta & Nitin Jain |
| 09:00 – 09:30 | Editing Models in the Data-free Regime— Chiranjib Bhattacharya | Chiranjib Bhattacharya |
| 09:30 – 10:00 | Dataflow Compatible End-to-End AI Acceleration— Sanchari Sen | Sanchari Sen |
| ☕ Break (10:00 – 10:30) | ||
| 10:30 – 11:00 | Building Effective Compilers for AI Accelerators— Uday Kumar Reddy | Uday Kumar Reddy |
| 11:00 – 11:30 | Cloud AI100 – Inference at Scale & SW Tool Chain for Accelerators (PyTorch / Triton)— Dibyendu Das | Dibyendu Das |
| 11:30 – 12:00 | Making LLM Inferencing Fast through Efficient KV Cache Update— Govindarajan Ramaswamy | Govindarajan Ramaswamy |
| 12:00 – 12:30 | RAMAN: An Edge AI Accelerator from High-Speed Imaging to Brain–Computer Interfaces— Chetan Singh Thakur | Chetan Singh Thakur |
| 🍽️ Lunch (12:30 – 13:10) | ||
| 13:10 – 13:40 | Distributed Inferencing— Subramanian Anantharaman | Subramanian Anantharaman |
| 13:40 – 14:10 | Inferencing on the Edge— Kapil Soni | Kapil Soni |
| 14:10 – 14:40 | AI Inferencing on Adreno GPU through Open-source Frameworks— Pavan Kumar | Pavan Kumar |
| ☕ Break (14:40 – 15:00) | ||
| 15:00 – 15:30 | Scheduling for LLM Inference Optimization— Parimal Parag | Parimal Parag |
| 15:30 – 16:30 | 💡 Brainstorming Session | |
Organised by
- IISc · ECE
- Distributed Systems Lab
- Qualcomm