Skip to content
View srishilesh's full-sized avatar
πŸ€–
πŸ€–

Highlights

  • Pro

Organizations

@fnplus @OpenMined @StreamrCVAT

Block or report srishilesh

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
srishilesh/README.md

Hey πŸ‘‹πŸ½, I'm Srishilesh P S

πŸ€– Lead AI Engineer β€’ πŸ› οΈ Product Strategist β€’ 🧠 Ex-ML Engineer

profile views


πŸ‘¨πŸ½β€πŸ’Ό About Me

  • πŸŽ“ CS + MBA (Masters' Union, Top 10%) | Former ML Engineer turned Lead AI Engineer & Product Strategist
  • πŸ”§ Currently architecting & deploying AI products at V-Mart Retail (NSE: VMART) β€” India's #1 value fashion omni-channel retailer with 500+ stores
  • πŸ€– 4.5+ years building enterprise-grade AI systems from 0 β†’ 1 in production β€” RAG, LLM workflows, MLOps, Document Intelligence
  • πŸ› οΈ Built Docupull β†’ 🌐 docupull.com | πŸš€ ProductHunt
  • πŸͺ„ Created Rugshak β†’ 🧢 Funky Rugs on Instagram
  • πŸ₯œ Founded Bambaii Foods β†’ 🌱 Healthy Snacking Brand

🧠 Things I've Worked On

  • 🏬 V-Mart Retail: Architecting RAG-based AI agents & 6 LLM-agentic workflows saving 20–50 manhours/month per process; built predictive optimization engines targeting 30% sell-through improvement across 500+ stores
  • πŸ” PyroAI @ Mr. Cooper Group (β†’ pyroai.com): Built enterprise RAG for 500+ call-center agents, reducing AHT by 37.5% with <200ms p95 latency; HITL annotation tool cutting TAT by 800% for 65+ annotators
  • πŸ“„ Docupull: AI-driven mortgage title verification processing 1,200+ orders/month at 98–100% SLA; reduced TAT from 24 hours to <1 minute
  • πŸ”Ž Community Compass: Hybrid RAG semantic search on pgvector indexing 3,000+ member profiles with <500ms end-to-end latency
  • 🀝 Contributor at OpenMined
  • πŸ“š Sharing thoughts on ML, startups & tech on Medium

πŸ’¬ Let's Talk About

  • 🧠 Applied ML & LLMs | πŸ€– Agentic Workflows & RAG | πŸ› οΈ Building from 0 β†’ 1
  • πŸš€ Technical Product Strategy | πŸ’» Scaling Internal AI Tools
  • πŸͺ„ Startups, Bootstrapping, Systems Thinking

πŸ› οΈ Tech Stack

ML & AI     β†’ RAG, DocumentAI, NLP, LLMs, MLOps, Computer Vision
Frameworks  β†’ PyTorch, TensorFlow, FastAPI, LangChain
Analytics   β†’ SQL, Excel, PowerBI, NoSQL
Cloud       β†’ GCP, Azure, AWS
Tools       β†’ Python, JavaScript, Java, Docker, n8n, Redis, PostgreSQL
Product     β†’ Strategy, A/B Testing, Roadmap, GTM

πŸ“« Reach Me


πŸ“Š GitHub Stats

Srishilesh's GitHub stats

Pinned Loading

  1. StreamrCVAT/streamrcvat StreamrCVAT/streamrcvat Public

    StreamrCVAT backend codes - contains ML models and other backend stuff

    Python 3

  2. fiesta fiesta Public

    Portal to view achievements of a student and keep in track of the latest events

    HTML 2 1

  3. PaySol PaySol Public

    PaySol is a chat based non-custodial crypto wallet that empower users to transact SOLs via P2P, keep track of financial records, advertise offerings, earn SPL tokens for every transaction.

    JavaScript 12 5

  4. Early-Fire-detection Early-Fire-detection Public

    Dynamic and Chromatic Analysis for Fire Detection and Alarm Raising Using Real-Time Video Analysis

    Python 13 3