Skip to content
View Avishkar-byte's full-sized avatar

Highlights

  • Pro

Block or report Avishkar-byte

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.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Avishkar-byte/README.md
Header
Typing Animation

B.Tech Electronics & Computer Engineering student at VIT Chennai, building at the intersection of autonomous systems, edge AI, and generative engineering platforms.

I engineer real-time perception pipelines for GPS-denied environments, deploy quantized CV models on resource-constrained edge hardware, and design end-to-end LLM systems that go from natural language input to physics-simulated, deployable artifacts. My work spans NVIDIA Jetson-based VIO stacks, on-device assistive wearables, medical device digital twin platforms, and full-stack AI web applications - each built to function under real hardware constraints and production latency requirements.

9 Utility Patents · 1 Design Patent · CVR 2025 Published · SIH '25 National Finalist · Tata Elxsi Top 60 / 47K+ · LTTS TECHgium® Top 7

Robotics & Perception

Visual-Inertial Odometry & state estimation ROS2 architecture & driver decoupling EKF sensor fusion at production Hz rates SLAM, loop closure & map resolution tuning Stereo-inertial calibration (D455 / RealSense) Edge deployment on Jetson Orin & Raspberry Pi Obstacle avoidance & reactive navigation

AI & LLM Systems

LLM requirement parsing & structured output RAG pipelines & agentic workflow design Groq SDK streaming inference integration Surrogate ML modeling on simulation telemetry Multi-physics solver orchestration (DAG-based) On-device CV: YOLO, MediaPipe, pose estimation Model quantization for real-time edge inference

Full-Stack & Systems

React 19 / Next.js 14 production applications FastAPI & Express.js backend architecture Real-time streaming APIs (chunked HTTP transfer) Docker containerization & split deployment Supabase (PostgreSQL) & in-memory session design 3D web rendering: Spline, Three.js, WebGL, Canvas MERN stack web application development

Python


React


Groq YOLO MediaPipe ROS2 SLAM EKF VIO VINS-Fusion Edge AI RAG


Recognition

LTTS Tata Elxsi SIH

Schrodinger Ecodeathon Ingenium


Research & Intellectual Property

9 Utility Patents Published across AI-driven systems, computer vision, and embedded technologies

1 Design Patent Published in robotics and assistive hardware systems

CVR 2025 - NIT Goa Research paper presented & published at the 5th International Conference on Computer Vision and Robotics

ORCID


Building

  • GPS-denied autonomous navigation for UAVs
  • On-device assistive AI systems at the edge
  • LLM-powered engineering automation pipelines
  • Production full-stack AI web applications

Exploring

  • MAVLink / PX4 pose injection from VIO output
  • ONNX export for cross-platform edge deployment
  • WebSocket-based real-time simulation streaming
  • FDA 21 CFR Part 11 compliance-aware AI design

SpectoV Assistive Technology · Product Development Intern
Python · YOLO · OpenCV · TensorFlow · Raspberry Pi
Built production ASL recognition pipeline (12,000+ images, 42 classes) · Deployed real-time CV inference at ~370ms latency via model quantization on Raspberry Pi


Stemtec AI and Robotics · Robotics & AI Intern
ROS2 · Gazebo · SLAM · EKF · Python · C++
Built STEMBOT autonomous navigation system · EKF sensor fusion at 15Hz reducing localization error 60% · SLAM at 92% loop closure accuracy · Obstacle avoidance 95% success rate


Algo Technologies Pvt. Ltd. · Software Developer Intern
MongoDB · Express.js · React.js · Node.js
Built and enhanced production MERN stack web applications · Designed RESTful API endpoints and reusable React component architecture · Contributed to scalable backend logic and MongoDB schema design


Portfolio GitHub LinkedIn ORCID Email

Pinned Loading

  1. SPARC-Smart-Perception-Assistive-Reality-Companion SPARC-Smart-Perception-Assistive-Reality-Companion Public

    This is a privacy-first, on-device wearable that enables real-time sign language translation, scene assistance, and safety monitoring. It integrates vision, IMU, and audio pipelines with a determin…

    Python

  2. Anukriti-Ai Anukriti-Ai Public

    An end-to-end generative system engineering platform that transforms natural language medical device requirements into physics-simulated, ML-surrogate digital twins. Combines LLM planning, numerica…

    TypeScript

  3. GradPath-AI GradPath-AI Public

    An AI-first study abroad companion for Indian students. Combines university matching across 55+ programs, conversational loan advisory, live SOP generation, and a gamified Dream Score - built on Gr…

    JavaScript 1

  4. HybridVIO-D455 HybridVIO-D455 Public

    Hybrid ROS + LibRealSense VIO pipeline for GPS-denied autonomous drone navigation. Decouples IMU and camera acquisition to resolve D455 USB driver contention, enabling centimeter-level VINS-Fusion …

    CMake