Thanks to visit codestin.com
Credit goes to github.com

Skip to content
View Jkanishkha0305's full-sized avatar
☯️
Yin-Yang
☯️
Yin-Yang

Organizations

@EddieHubCommunity @TEAM-IFOR

Block or report Jkanishkha0305

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
Jkanishkha0305/README.md

Neon header banner for J. Kanishkha

Hi Waving hand, I'm J. Kanishkha

AI Engineer @ Toyota Β· MSDS @ NYU CDS Β· MLE @ OpenHealthcare US

Turning research ideas into production-grade AI products across LLM systems, multimodal AI, evaluation, and cloud MLOps.

Animated coding workspace

Portfolio LinkedIn Hugging Face Google Scholar Kaggle Email

Profile views GitHub followers Location New York Focus on LLM Systems and MLOps

What I Build πŸ› οΈ

  • LLM systems: RAG and agent workflows, evaluation and guardrails, retrieval orchestration, and production deployment
  • Applied AI: NLP, transformers, multimodal applications, computer vision pipelines, and human-centered products
  • ML engineering: MLOps, model serving, observability, cloud-native tooling, and research-to-production delivery

Current Focus 🎯

  • Shipping practical LLM systems and AI agents for real users
  • Going deeper on retrieval, orchestration, evaluation, and observability
  • Studying from-scratch model implementations to sharpen fundamentals
  • Building AI products across healthcare, enterprise automation, and mobility

Ask Me About πŸ’¬

RAG pipelines Agent workflows Transformers Multimodal AI MLOps Graph reasoning

Featured Projects πŸš€

Systems And Products

  • mcp-automation-system: Enterprise AI automation for PR review, Asana context, and Slack delivery
  • Medibot-GraphRAG: Hospital-focused GraphRAG assistant built with LangChain and Neo4j
  • Theramind: Offline mental-health app powered by a fine-tuned small language model

Research And Learning

  • Align2Act: Instruction-tuned models for human-aligned autonomous driving
  • fashion-recsys: End-to-end recommendation system for fashion retrieval and ranking
  • LLMs-from-Scratch: Hands-on implementations of LLMs, SLMs, and VLMs for experimentation

Technical Stack ⚑

Languages And Frameworks

Python SQL C++ TypeScript React Next.js Node.js React Native FastAPI FastMCP

ML And AI

PyTorch TensorFlow Scikit-learn Spark Hugging Face LangChain LangGraph Google ADK ONNX vLLM Ray

MLOps And Cloud

AWS Google Cloud Databricks Docker Kubernetes MLflow OpenTelemetry Grafana Prometheus Jaeger DVC DagsHub GitHub Actions LangSmith Opik Weights and Biases

Data, Retrieval, And Systems

MongoDB PostgreSQL Pinecone Qdrant Neo4j OpenSearch Redis Kafka CUDA Git

Signal Board ✨

Research to production Primary themes Open to collaboration and applied AI

3D Activity 🌌

3D contribution landscape

GitHub Signals πŸ“Š

Repositories per language Most commit language

GitHub streak stats

Contribution Snake 🐍

GitHub contribution snake animation

If you're building with LLMs, applied ML, or production AI systems, let's connect.

Pinned Loading

  1. parallax parallax Public

    EmpireHack '26 - Finalist πŸ…| Parallax - AI persona agents that test your product from 5 different perspectives just like a human with emotions

    TypeScript 1

  2. mcp-pr-reviewer mcp-pr-reviewer Public

    πŸ€– Enterprise AI automation system using Model Context Protocol (MCP). Intelligent PR reviewer that analyzes GitHub pull requests, fetches context from Asana, and posts actionable insights to Slack.…

    Python 1

  3. Medibot-GraphRAG Medibot-GraphRAG Public

    GraphRag based chatbot built using LangChain and Neo4j, designed for hospital systems. The chatbot retrieves structured and unstructured data about patients, visits, physicians, insurance payers, …

    Python 6

  4. dungeon-agents dungeon-agents Public

    Two AI agents collaborate and explore a fog-of-war dungeon while reasoning every move with agent observability.

    Python 1

  5. fashion-recsys fashion-recsys Public

    End-to-end fashion recommender system using collaborative filtering and matrix factorization

    Jupyter Notebook 3

  6. LLMs-from-Scratch LLMs-from-Scratch Public

    A curated collection of Large Language Models(LLMs), Small Language Models(SLM), Visiona Language Models(VLM) implemented from scratch for Learning, experimentation, and innovation across Text, Vis…

    Jupyter Notebook