Skip to content
View ajinkyapuar's full-sized avatar

Block or report ajinkyapuar

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. 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
ajinkyapuar/README.md

Ajinkya Puar

Building intelligent systems at the intersection of AI, Computer Vision & Human-Computer Interaction

AI Systems · Computer Vision · Agentic Software · Eye Tracking · Edge AI · Spatial Computing

LinkedIn GitHub EyelignAI Email


About

I'm a software engineer, technical lead, and R&D builder focused on turning emerging technologies into real-world products.

My work spans artificial intelligence, computer vision, agentic systems, mobile/edge AI, cloud infrastructure, extended reality, and human-computer interaction.

I enjoy working across the full stack of a difficult technical problem:

Research → Architecture → Prototype → Product → Deployment

Much of my current work explores how AI systems can perceive, reason, act, and adapt to people in real time.


What I'm Building

👁️ EyelignAI — Vision Intelligence

Building camera-based gaze intelligence that turns standard RGB cameras into real-time eye-tracking systems.

  • Monocular gaze estimation
  • On-device computer vision
  • Gaze analytics & visualization
  • Mobile / Android SDKs
  • Accessibility & gaze interaction
  • Attention-aware applications
  • Eye-health and research tooling

→ eyelign.ai

🤖 Agentic AI Systems

Exploring and building systems where AI agents can move beyond chat interfaces and actually reason over data, use tools, operate software, generate artifacts, and complete multi-step workflows.

Current areas of interest include:

  • Autonomous AI agents
  • Multi-agent orchestration
  • Computer-use agents
  • Local / private AI
  • MCP & tool-use architectures
  • Agentic data analysis
  • AI-generated interfaces
  • Human-in-the-loop systems

🧠 Applied AI & Computer Vision

My broader work includes:

  • Object detection & instance segmentation
  • Image and video understanding
  • Camera calibration
  • Facial recognition
  • Multimodal AI
  • GPU inference systems
  • Visual analytics
  • Edge AI deployment

🥽 Spatial & Intelligent Computing

I've also built systems across:

  • AR / VR / MR
  • Spatial interfaces
  • AIoT
  • Real-time communication
  • Simulation
  • HMD platforms
  • Cloud-connected devices

Selected Work

Python + OpenCV tooling for camera calibration, chessboard detection, image capture, calibration dataset management, and intrinsic camera parameter estimation.

🥽 EdificeVR

Immersive interior-design and commerce experiences combining real-time 3D, VR, product visualization, and interactive environments.

🎥 TENS — Tag Enforced Neural System

Computer-vision system for large-scale video understanding and audio-visual search, including classifiers, keyframe extraction, shot-change detection, and GPU-based processing.

👤 SnapClass

Computer-vision attendance system using facial recognition and traditional machine-learning pipelines.


Technology

AI / Vision

Python · PyTorch · TensorFlow · Keras · OpenCV · scikit-learn · NumPy · Pandas

Agentic AI

LLMs · AI Agents · MCP · Tool Use · Multi-Agent Systems · Local LLMs · Ollama

Backend / Systems

Flask · Node.js · REST APIs · WebSockets · RabbitMQ · Celery · Docker · Linux

Mobile / Edge

Android · Java · Kotlin · C++ · On-device AI

Cloud / Infrastructure

AWS · Azure · AliCloud · GPU Infrastructure · Docker

Spatial Computing

Unity3D · AR · VR · MR · Computer Vision · IoT


Areas I'm Exploring

AI Agents
├── Multi-agent systems
├── Computer use
├── Autonomous workflows
├── Local AI
└── Human-in-the-loop systems

Vision Intelligence
├── Eye tracking
├── Gaze estimation
├── Ocular imaging
├── Multimodal perception
└── On-device inference

Intelligent Interfaces
├── Spatial computing
├── Accessibility
├── Adaptive interfaces
└── Human-computer interaction

Philosophy

Build systems that don't just process information —
build systems that can perceive, reason, interact, and adapt.

I’m particularly interested in problems that sit between research and engineering, where new AI capabilities need to become reliable, deployable products.


Let's build something intelligent.

LinkedIn · EyelignAI · GitHub · Email

@ajinkyapuar's activity is private