| FazBrowse GitHub Viewer | Trending | | Home |
| Tools: [Original HTTPS Page] |
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.
Contact GitHub support about this userβs behavior. Learn more about reporting abuse.
Report abusename: Sudharshan Ramesh
location: New York, United States
education: M.S. Computer Engineering @ New York University
roles:
- AI/ML Engineer
- Robotics Enthusiast
- Full Stack Developer
currently_building:
- MCP agents for automated research workflows
- Reinforcement learning for autonomous drones
- Collaborative real-time coding platforms
interests:
- Making robots that don't fall over
- Breaking neural networks (for science)
- Building things that scale from 0 β millions|
Autonomous drone obstacle avoidance using PPO & MPC β NYU Reinforcement Learning |
ποΈ 3D Urban Scene Reconstruction2-pass pipeline for 3D reconstruction of urban environments |
π€ OpenBenchPublisherMCP agent for automatic dataset creation, research paper analysis & prototyping |
π» Meet2Code V2Collaborative coding platform with video chat + git integration β rebuilt with modern stack |
π·οΈ Hexapod Robot18-DOF hexapod robot β Arduino + Raspberry Pi integration |
π¦ MLOps β BirdCLEFBird species prediction from acoustic sensor data in wildlife environments |
| Project | Description | Tech |
|---|---|---|
| RL_PPO_DroneObjectAvoidance | Autonomous drone obstacle avoidance using PPO & MPC | RL PPO MPC |
| RL_DroneLooping | Drone aerobatic maneuvers via non-linear optimal control | RL MPC Control |
| OpenBenchPublisher | MCP agent for automated dataset creation & research analysis | MCP LLM Agents |
| Roberta-Fine-tuning-with-LoRA | Fine-tuning RoBERTa with PEFT/LoRA on AG News | LoRA PEFT NLP |
| Siamese-Network | Face detection pipeline on CelebA using Siamese Networks | CV PyTorch |
| JailBreaking-ResNet-34 | Adversarial attacks on ResNet-34 | Adversarial ML CV |
| DeepLearning-CIFAR-10 | CIFAR-10 classification with ResNets under 5M params | ResNet CV |
| SAM2-HandMask | Hand segmentation with SAM2 + MediaPipe | SAM2 MediaPipe |
| MLOps | BirdCLEF: Bird species prediction from acoustic sensor data | MLOps Audio ML |
| DeBERT-Paper-Order-segmentation | Paper section ordering with DeBERTa | NLP Transformers |
| GenAI | RAG-based generative AI chatbot | RAG LLM GenAI |
| Project | Description | Tech |
|---|---|---|
| Hexapod | 18-DOF hexapod robot with Arduino + Raspberry Pi | C++ Arduino RPi |
| IAMTCQ | 4-legged crawler robot for terrain navigation | Robotics C++ |
| Embedded-Sentry | Gesture-based unlock system on STM32F429I via gyroscope | STM32 RTOS C |
| Project | Description | Tech |
|---|---|---|
| Meet2Code_V2 | Collaborative coding platform with video + git (v2) | React WebRTC CRDT |
| Meet2Code | Real-time collaborative code editor with communication | JS Yjs PeerJS |
| CampusConnect | Campus networking app with Spring Boot backend | Spring Boot Vue.js |
| InternetBanking-Application | Online banking app illustrating OOAD principles | React Express MySQL |
| energy-manager-app | Energy management dashboard β DevFest '25 @ Columbia | React IoT AI |
| groq-gen-ai | Real-time energy consumption prediction with Groq + EIA data | Groq GenAI |
| StreetAwareApp | Urban data collection app for StreetAware project | IoT Mobile |
A project done as part of Reinforcement Learning Lecture at New York University leveraging PPO and MPC.
Jupyter Notebook 2
Fine tuning a Roberta model with peft and lora for AGNEWS dataset
Jupyter Notebook
Developed a Hexapod robot as an IFP. Works using an arduino integrated with raspberry PI with 18 DOF
C++
| Back | FazBrowse Home | New Git URL |