Seshadri User

I’m a Software Engineer II with strong foundations in Go, distributed systems, and machine learning. I build production-grade, low-latency backend services and ML-driven systems, with a focus on performance, reliability, and end-to-end delivery across real-time architectures, MLOps, and data pipelines.

At Techmojo Solutions Pvt Ltd, I led a team to architect and implement a unified feed entity microservice that improved feed availability to 99.99% and eliminated duplicate records across systems processing 1B+ events. I helped drive critical design choices by authoring technical design docs and presenting consistency vs. query-latency tradeoffs to stakeholders.

I’ve also optimized high-throughput event-driven platforms handling billion-scale Kafka events—reducing storage costs by 40% and scaling REST APIs and internal services to improve platform integration velocity by 20%. During peak traffic incidents, I’ve personally diagnosed and resolved 1B+ message backlogs within hours under SLA pressure, and I built and owned Prometheus/Grafana monitoring with runbooks and post-incident reviews.

Beyond backend engineering, I develop ML-oriented systems—building an LLM routing and model lifecycle platform in Python using FastAPI, PyTorch, and Redis, including contextual bandit routing (Thompson Sampling) with <5ms routing latency per request and stabilizing end-to-end request success to 100%. I’m motivated by systems ownership, pragmatic engineering, and delivering measurable impact.