Sharath C.

About Sharath C.

Lead Software Engineer with 8.5+ years of experience building and scaling low-latency, high-throughput backend and ML systems. Developed large-scale experimentation platforms, ML serving layers, and revenue-critical travel systems at global scale, serving millions of travellers worldwide. Founded and operationalised org-wide GenAI infrastructure, including A100/H100 GPU clusters in Azure cloud, distributed LLM inference, and centralised model serving platforms. Strong background in distributed systems design, low-latency APIs, and scalable ML/LLM deployment.

Experience

1. Lead Engineer

   MakeMyTrip

   Feb 2020 — Present · Bengaluru, IN

   Core Tech Stack: Python, C++, Java | vLLM, Triton/TensorRT, Llama-Factory, GraphRAG | FastAPI, Docker, AWS/Azure (A100/H100), Qdrant, Aerospike, Redis.Key Achievements: LLM Infrastructure & GovernanceArchitected and deployed an org-wide LLM serving platform powering 20+ product teams, handling 750K+ daily chat completion requests.Optimized high-throughput inference using vLLM and Triton/TensorRT on A100/H100 clusters with MIG partitions to power the Myra AI chatbot.Designed a centralized AI Gateway with unified authentication and rate-limiting, governing internal fine-tuned models and external OpenAI/Gemini API usage across the company. Revenue-Critical Product EngineeringBuilt the \"Flights PriceLock\" backend from scratch, a travel fintech product allowing users to lock fares for a premium.Scaled to 12M searches per day with a p95 latency of ~200ms, generating $800K (INR 6–8 Cr) in annual revenue.Implemented backend services for the Odin Multi-Armed Bandit (MAB) platform, serving 6M requests daily at sub-50ms latency for cross-vertical experimentation (Flights, Hotels, Insurance). Multi-Modal AI & Vision PipelinesEngineered a large-scale vision pipeline on Azure H100 clusters to upscale 100K+ images to 4K resolution.Integrated a dual-model audit layer (Qwen VLMs & GPT-4o) to ensure visual integrity and filter synthetic textures for strict SEO quality standards.Deployed production-grade AI solutions including post-sales Speech-to-Text with LLM recommendations, a vLLM-based TTS streaming service for Indic accents, and CV-based cab luggage fitting analysis. MLOps & Data EngineeringManaged end-to-end ML lifecycles using Airflow, S3, and Databricks.Built scalable vector search and caching layers using Qdrant, Aerospike, and ElasticSearch.Established full-stack observability via Grafana, Scalyr, and Kafka.

Education

• NIIT University

  Bachelor’s Degree, Computer Science and Engineering

  2013 — 2017

• Sacred Heart Public School

  Higher Secondary, Secondary School

  2010 — 2011

• Nava nirman Public School

  Computer science with Maths, Senior Secondary School

  2012 — 2013