KEDAR SATHE
@wtfkedar
I'm Kedar, an AI Engineer and SDE from Pune. I build AI systems, agentic pipelines, and infrastructure — and I like understanding things at a deep level before building them.
Most of my time goes into AI research, model building, and finetuning — figuring out how models think, where they break, and how to make them actually useful. I'm also into agentic systems: giving LLMs tools, memory, and the ability to reason through problems end-to-end.
On the other side, I enjoy low-level work — kernel modules, building libraries, writing node packages. There's something satisfying about working close to the metal.
Open to good work and interesting conversations — feel free to reach out.
Work Experience
Unique School App India
Software Developer Intern
- Led React Native cross-platform app migration for the school management platform
- Built and maintained backend services and REST API endpoints
- Set up and improved CI/CD DevOps pipelines for automated deployment
- Collaborated on system architecture to improve scalability and reliability
Jan 2026 – Apr 2026
Pune, India
Mermaid.js
Open-Source Maintainer
- Maintains Mermaid.js, a widely used OSS diagramming library
- Triages issues, reviews PRs, and guides contributors
- Resolved 8+ bugs affecting rendering fidelity and cross-platform compatibility
- Shipped PRs with new functionality and performance improvements
2024 – Present
Remote
Freelance
ML Engineer & Full-Stack Developer
- Built responsive web applications using Next.js, React, and TailwindCSS
- Implemented ML pipelines and deployed models on cloud platforms
- Worked with clients to deliver intuitive user experiences on time
2023 – Present
Remote
Tech Stack
This list grows faster than my training loss curves — and I love that.
< Languages />
< ML & AI />
< Frameworks & Frontend />
< Backend & Database />
< DevOps & Cloud />
< Developer Tools />
Proof of Work
Things I built when curiosity got the better of me.
X-Bird
Chrome extension for AI-generated contextual replies on X.com. Supports multiple tones: Auto, Savage, Supportive, Funny, Professional, Casual — with adjustable intensity control.
LocalRaG
Retrieval-Augmented Generation app with Streamlit UI, PostgreSQL vector storage via PgVector, HNSW indexing, and LLaMA 3 (70B). Response latency under 1 second for most queries.
Blogs & Notes
I write about AI experiments, system design breakdowns, and lessons from building in public. Check out my active blog for technical deep-dives and dev notes.
neuronreads.vercel.app
AI, ML systems, dev notes & more →