Skip to content
View rishaviitd's full-sized avatar
😄
😄

Block or report rishaviitd

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 is supported. This note will only be visible to you.
Report abuse

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

Report abuse
rishaviitd/README.md

Hi 👋, I'm Rishav Kumar | IIT DELHI '25

I am currently a Data Scientist with deep expertise in AI/ML, data pipelines, backend engineering, and distributed systems. I thrive on crunching numbers and crafting innovative solutions—taking complex ideas from initial prototyping and experimentation all the way through to deploying them as scalable, production-ready systems.

Coding

📝 Visit my website: nirmata.dev
📫 Email me at: rishav01iitd@gmail.com

🎯 THINGS WHERE I HAVE WORKED AROUND SO FAR

🤖 AI AGENT

🌐 TRANSFORMERS WORLD

  • State-of-the-Art (SOTA) Transformer architectures in Vision, Speech, and Text
  • Transformer inference optimizations (e.g., vLLM)
  • SFT (Supervised Fine-Tuning) for embedding models, rerankers, and LLMs

🎮 REINFORCEMENT LEARNING

🛠️ DATA ENGINEERING

  • Architecting robust data engineering pipelines
  • Advanced database design and modeling

📊 DATA SCIENCE

  • Supervised & Unsupervised Learning
  • Recommender Systems
  • Time Series Analysis

Popular repositories Loading

  1. kaggle.rental.product.recommendation kaggle.rental.product.recommendation Public

    The task was to build a model that, given the metadata of a browsing session and the sequence of visited pages, predicts which product pages the user will visit next. Achieved best score on Kaggle

    Python 1

  2. rishaviitd rishaviitd Public

  3. better.call.ai better.call.ai Public

    Production-grade agentic legal RAG stack using Next.js, FastAPI, LangGraph, Temporal, Postgres, Redis, Qdrant, OpenMemory, Zembed, and LlamaParse — containerized with Docker, deployed on AWS (ECS, …

  4. kaggle.orbit.wars.game.agent kaggle.orbit.wars.game.agent Public

    The goal of this competition was to create and/or train AI bots to play a novel multi-agent 1v1 or 4p FFA game against other submitted agents.

    Jupyter Notebook

  5. kaggle.toxic.comment.classification.challenge kaggle.toxic.comment.classification.challenge Public

    Classifying text comments into 6 different labels as a multi class classification problem