🌙
Nilesh Sarkar
/
Blog
Thoughts and project notes on AI, LLMs, and Engineering.
Back to Home
Events
Research & Engineering
LectureToSlides: Video to PDF Converter
Tool • Computer Vision & AI
A browser-based tool that uses computer vision to detect slide transitions in lecture videos and Gemini AI to generate summaries and quizzes.
Large Language Model Architecture & Compression Research
Jan 2025 • Research Project
Independent research on efficient LLM architectures and compression for constrained, real-world deployments; training compact transformers and studying compression as an analytical tool.
Sim-to-Real Humanoid Control
Ongoing • Research Project
Developing robust locomotion policies for a miniature humanoid robot using Deep Reinforcement Learning (PPO) in NVIDIA Isaac Sim.
AIR Research Group: Autonomous Surveillance Drone
Present • Research Group
Building a robust, autonomous drone system capable of performing real-time surveillance tasks with the AIR Research Group.
Local AI Connect: Private Copilot for VS Code
Dec 2025 • VS Code Extension
A comprehensive guide to my new VS Code extension that bridges local LLMs (Ollama, LM Studio) directly into the editor for offline, privacy-first coding assistance.
Events & Talks
Bangalore Drone Community Meetup
Oct 2025 • Event
Insights from the Bangalore Drone Community Meetup: Startups, regulations, and the future of UAVs.
Build with AI: Google’s Agent Development Kit & MCP
May 2025 • Workshop
My experience at the Build with AI workshop organized by Google and Deep Tech Stars. Exploring ADK and Model Context Protocol.
NASA Space Apps: Landsat SR on the fly
Oct 2024 • Hackathon
A solution for comparing ground observations with Landsat Surface Reflectance data and alerting users of satellite overpasses.
Notes & Experiments
Sutskever's List: Foundational AI Research
Reading List • Ilya Sutskever
A curated collection of 30 research papers that Ilya Sutskever (OpenAI) claims contain "90% of what matters today" in AI.
Hacker's Guide to Neural Networks
Reproduction • Andrej Karpathy
A reproduction of Andrej Karpathy's legendary "Hacker's Guide to Neural Networks". A code-first approach to understanding backpropagation and real-valued circuits.
Agentic RAG chatbot with LangGraph — project notes
Jul 2024 • Project Notes
Building an Agentic RAG workflow with LangGraph and Streamlit. The system prioritizes PDF Q&A and falls back to web search when necessary.
Image generation with Stable Diffusion — quick notes
Apr 2024 • Experiments
Running experiments with Stable Diffusion v1.5 via diffusers. Exploring negative prompts, guidance scales, and prompt engineering basics.