Sudarshan Srinivasan
Deep Learning System Architect, Large Language Models/Deep Learning/GPU Architecture
- Role
- Research Scientist Ai Ml at Intel Labs
- Location
- Bengaluru, IN
- LinkedIn followers
- 500 followers
Experience
Research Scientist Ai Ml
Aug 2016 — Present
Deep learning Research ScientistWork on System architecture (focus on deep learning workloads) design for Xe line of datacenter discrete gpus1) Involved in developing/leading the datacenter System performance models for AI WL (Specifically LLM) for distributed training/Inference at scale for DL workloads for candidate Xe-GPU. Model will consider Key factors that affect application performance for a single node and beyond to multi-node clusters.2) System Design for next generation workloads in Large Model NLP Training, Recommendation Models. This work led directly into the planning and development of Intel LLM based applications across functional teams and customer engagements. Defined a long-term strategy for evaluating node/rack design choices for an efficient deployments. 3) Performance correlation with Post silicon data and also detailed profiling/correlation with competitive GPU systems
Education
University of Massachusetts, Amherst
Doctor of Philosophy (Ph.D.)
2012 — 2016
University of Massachusetts Amherst
Masters
2009 — 2011
Vellore Institute of Technology
B-TECH
2005 — 2009
Find verified contacts for anyone on LinkedIn
Unifers gives sales teams verified emails and direct dials, enriched profiles, and outreach that lands in the inbox.
Free plan included · No credit card required
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.