Summary
Devang Sharma is a Senior Software Engineer and AI specialist with 8+ years building production-grade distributed systems, from low-latency inference pipelines to scalable microservices across AWS, GCP, and Azure. He architects multi-agent frameworks and optimizes transformer internals—attention, tokenization, quantized inference and kernel fusion—to deliver reliable, high-throughput LLM deployments on GPU/TPU clusters. His background spans full-stack and backend leadership at Meta, Amazon, BMO and Grab, where he pushed migrations to cloud, hardened security/compliance, and achieved sub-100ms P90 latencies for mission-critical APIs. Devang blends research-grade model evaluation and fine-tuning (LoRA/PEFT/QLoRA) with pragmatic engineering—instrumenting observability and resilience for agentic systems under production load. He pairs strong systems fundamentals in C/C++, Go and Python with developer-facing tooling and API design, and mentors engineers across industry programs. Based in Toronto, he’s driven by turning advanced AI research into auditable, high-impact products used at scale.
9 years of coding experience
7 years of employment as a software developer
Indian Institute of Technology Delhi (IIT Delhi)
BITS Pilani, Birla Institute of Technology and Science
Kendriya Vidyalaya Sangathan
Bachelor of Technology ( B-Tech), Computer Science Engineering, Bachelor of Technology ( B-Tech), Computer Science Engineering at Guru Gobind Singh Indraprastha University
English, Hindi, German, Spanish, Italian