Skip to content
View harshitworkmain's full-sized avatar
💭
hey let's build
💭
hey let's build
2 followers · 10 following

Highlights

Block or report harshitworkmain

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
harshitworkmain/README.md

Harshit Singh

Engineering intelligent, hardware-backed AI systems that bridge the physical and digital worlds.

I'm a pre-final year Electronics and Computer Engineering undergrad specializing in Computer Vision, Machine Learning, and applied AI systems. I build end-to-end intelligence pipelines—ranging from real-time deep learning recognition models to autonomous robotic navigation nodes. I focus on developing production-ready inference architectures with measurable real-world impact.

⚡ Technical Arsenal

Programming Languages
Python C++ C Java TypeScript JavaScript HTML CSS

AI & Robotics Frameworks
PyTorch TensorFlow OpenCV ROS 2

Libraries: YOLOv8, MediaPipe, NumPy, Pandas, CVZone


Full-Stack & Cloud
React Node.js Express SQLite Flask

Embedded Systems & Edge Hardware
Arduino Raspberry Pi Linux

Systems & Development Tools
VS Code MATLAB GitHub Git Docker

🚀 Featured Engineering Feats

Project Focus Impact & Metrics
Indoor Autonomous Mobile Robot ROS 2, LiDAR Developed a ROS 2 Jazzy reactive navigation node processing LiDAR /scan data at 10 Hz. Enabled real-time obstacle evasion, reducing indoor collision events by ~80%.
SPARC System OpenCV, Deep Learning Engineered a real-time ISL/ASL recognition pipeline using MediaPipe and TF/Keras. Achieved 92.4% accuracy at 15–20 FPS with end-to-end latency under 180 ms.
TRIDENT Ecosystem IoT, Robotics, Orchestrator Built an edge-based disaster rescue platform streaming telemetry with >95% uptime. Automated probabilistic priority scoring, reducing manual triage time by 40%.
DrishtiGuide IoT Edge AI, YOLOv8 Deployed a multimodal navigation pipeline integrating YOLOv8 and ESP8266, achieving sub-120 ms inference latency and reducing missed-obstacle events by ~60%.
NeuroBridge AI Full-Stack ML, Telehealth Engineered a continuous remote care ecosystem with 468 facial landmarks tracking via MediaPipe. Built interactive clinical dashboards supported by Node 22 & React 19.

🏆 Experience & Achievements

  • AI & Robotics Intern @ STEMTEC: Developed ROS 2 Jazzy navigation nodes processing LiDAR data at 10 Hz, reducing collision events by ~80% during indoor testing.
  • National Finalist @ SIH 2025: Advanced to the Top 5 out of 500 teams nationally.
  • Top Honors: Winner of Innovate X Impact (V-NEST) and 3rd Place out of 50+ teams at Technovation 2025 (₹15k).
  • Leadership: Deputy Operations Lead for Team Namo Nirvana, managing a 100+ member team and boosting execution efficiency by 35%.

📈 GitHub Analytics

Harshit's GitHub Stats Harshit's GitHub Streak

Pinned Loading

  1. trident trident Public

    This is an integrated disaster rescue ecosystem for real-time situational awareness and priority classification. It integrates edge wearable telemetry (ESP32, MAX30102, GSR, MPU6050, NEO-6M GPS) an…

    Python 1 1

  2. drishtiguide drishtiguide Public

    IoT-based, multi-node assistive ecosystem (RPi 4, ESP32 DevKit V1, ESP8266 NodeMCU) for real-time obstacle detection, spatial haptics & safety monitoring. Integrates ultrasonic, IMU & GPS data via …

    Python 1

  3. weather-risk-ai-lstm weather-risk-ai-lstm Public

    An advanced AI system leveraging Multivariate LSTM networks for hyper-local weather forecasting and Graph Theory to simulate urban risk propagation, flood capacities, and storm trajectories.

    Jupyter Notebook

  4. Neurobridge Neurobridge Public

    Forked from Akshayk1129/Neurobridge

    Open-source remote care platform for ASD. Built with React and Node.js, featuring MediaPipe AI screening, Phaser 3 therapy games, Socket.IO teleconsultation, and clinician analytics dashboards.

    JavaScript