Skip to content
View IshaanKalra2103's full-sized avatar

Block or report IshaanKalra2103

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
IshaanKalra2103/README.md

Hi, I'm Ishaan πŸ‘‹

Machine Learning Researcher β€’ Full-Stack Engineer β€’ Builder


πŸ‘¨β€πŸ’» About Me

I build intelligent systems that solve real problems. My work spans machine learning research, full-stack development, and entrepreneurship. I'm drawn to challenges at the intersection of AI, policy research, and scalable product development. I focus on shipping tools that are both technically sophisticated and genuinely useful.


πŸš€ Some things I am building

🎡 VYBZ – Music Discovery Platform

A cross-platform startup connecting people through music discovery. Building web (Next.js, GSAP, Tailwind, Supabase) and mobile (Flutter) experiences.

Next.js Flutter Supabase

🧠 SciKernel – Autonomous CUDA Optimization

Built an agentic system using LangGraph that autonomously optimizes CUDA kernels. Exploring how AI can improve low-level performance optimization workflows.

Python CUDA LangGraph

πŸ“š Lit2Chem – Literature to Knowledge Pipeline

Automated system that converts chemistry literature into structured knowledge graphs using fine-tuned LLaMA models. Bridges the gap between research papers and actionable data.

PyTorch LangChain

🌍 ML for Policy Research – Center for Global Sustainability

Leading ML team exploring domain-specific LLM use cases for environmental policy. Adapting AI agents with reinforcement learning optimizers (GRPO, MIPROv2) to support data-driven policy decisions.

TensorFlow JAX


πŸ”§ Full Stack + ML Engineering

Languages
Python β€’ JavaScript β€’ TypeScript β€’ C++ β€’ Rust β€’ R β€’ Java

ML/AI
PyTorch β€’ TensorFlow β€’ JAX β€’ LangChain β€’ LangGraph β€’ RAG β€’ Fine-tuning β€’ RLHF

Frontend
React β€’ Next.js β€’ Flutter β€’ Tailwind β€’ GSAP

Backend & Cloud
Flask β€’ Node.js β€’ FastAPI β€’ Supabase β€’ Firebase β€’ GCP β€’ AWS

Systems & Tools
CUDA β€’ Distributed Systems β€’ Git β€’ Docker


πŸ“š What I'm learning right now

learning_goals = {
    "AI/ML": [
        "Reinforcement learning for agent optimization",
        "Fine-tuning LLMs for domain-specific tasks",
        "Scaling distributed training workflows",
        "Agentic systems with LangGraph"
    ],
    "Systems": [
        "CUDA kernel optimization",
        "Distributed systems architecture",
        "Performance engineering at scale"
    ],
    "Product": [
        "Building 0β†’1 consumer products",
        "Cross-platform mobile development",
        "Growth and user retention strategies"
    ],
    "Research": [
        "Code retrieval and semantic search",
        "Policy-oriented ML applications",
        "NLP for scientific literature"
    ]
}

πŸŽ“ About me

Education

  • Computer Science at University of Maryland
  • Minor in Statistics

Philosophy

  • I build systems that bridge research and production
  • I ship quickly, iterate constantly, and optimize relentlessly
  • I care about making AI useful, not just impressive

LinkedIn GitHub

Feel free to reach out anytime.


"Build systems that matter. Ship things that work. Learn relentlessly."

Always building. Always learning. Always shipping.

Pinned Loading

  1. lit2chem lit2chem Public

    Python

  2. umdfiresa2023/fire-summit-whistle-blowers umdfiresa2023/fire-summit-whistle-blowers Public

    JavaScript