Software engineer in Bangalore. Building real-time AI systems — voice pipelines, streaming inference, LLM orchestration at Jobtwine.

Currently interested in inference optimization, speculative decoding, and small-model routing for latency-bound systems.

Writing → medium.com/@OmsharmaOfficial Code → github.com/justomsharma