Computer Vision & AI Software Developer

Role & responsibilities

Job Title: Computer Vision & AI Software Developer

Experience: 2+ Years

Location: Warje, Pune, Maharashtra 411058

Employment Type: Full-Time

Work Mode: Work from Office

Job Summary

We are looking for a hands-on Computer Vision & AI Software Developer with 2+ years of practical experience in developing, deploying, and optimizing AI-based computer vision and real-time video analytics solutions.

The ideal candidate should have strong expertise in Python, OpenCV, PyTorch/TensorFlow, YOLO, real-time video processing, object detection/tracking, RTSP/IP cameras, FastAPI/Flask, databases, Docker, and Edge AI platforms such as NVIDIA Jetson and Raspberry Pi.

This is not a generic AI/ML or web development role. Candidates must have hands-on experience building and deploying real-time Computer Vision applications.

Key Responsibilities

• Design, develop, test, and deploy AI/Computer Vision software solutions.
• Develop real-time video analytics applications using Python, OpenCV, NumPy, and AI/Deep Learning frameworks.
• Implement object detection, object tracking, bounding boxes, ROI/polygon zones, motion analysis, and spatial/temporal analysis.
• Work with YOLO or equivalent object detection models, including training, fine-tuning, evaluation, and optimization.
• Develop solutions for multiple IP camera/RTSP streams and real-time video processing.
• Implement video buffering, pre-event/post-event recording, automatic stream reconnection, and video storage.
• Develop AI event detection logic using Track IDs, zone-based detection, state machines/rule engines, confidence scoring, and alert filtering.
• Optimize models and applications to reduce false positives/false negatives and improve real-time inference performance.
• Develop REST APIs using FastAPI, Flask, or equivalent frameworks.
• Work with WebSocket, JSON, API integrations, authentication, and real-time data communication.
• Design and manage databases using PostgreSQL/MySQL and SQL.
• Integrate AI solutions with web dashboards using JavaScript, HTML/CSS, React, or equivalent technologies.
• Deploy applications using Docker, Docker Compose, Linux/Ubuntu, and NVIDIA GPU/CUDA environments.
• Develop and deploy AI solutions on Raspberry Pi and NVIDIA Jetson platforms.
• Integrate cameras, sensors, and IoT devices with AI/Computer Vision systems.
• Debug, troubleshoot, and optimize applications for performance, reliability, and scalability.

Must-Have Skills

Programming

• Advanced Python
• Object-Oriented Programming
• Multithreading / Multiprocessing
• Data Structures & Algorithms
• Git / GitHub
• Debugging and performance optimization

Computer Vision

• OpenCV
• NumPy
• Image & Video Processing
• Object Detection & Tracking
• Bounding Boxes
• ROI / Polygon Zones
• Motion Analysis
• Spatial & Temporal Analysis

AI / Deep Learning

• PyTorch or TensorFlow
• YOLO or equivalent object detection frameworks
• Model training / fine-tuning
• Dataset preparation and annotation
• Model evaluation
• Precision / Recall / mAP
• False-positive and false-negative reduction
• Real-time inference optimization

Real-Time Video Analytics

• IP Cameras
• RTSP
• FFmpeg
• Multiple camera streams
• Video buffering
• Pre-event / Post-event recording
• Automatic stream reconnection
• Video storage and retrieval

Backend

• FastAPI / Flask
• REST APIs
• WebSocket
• JSON
• API integration
• Authentication

Database

• PostgreSQL / MySQL
• SQL
• Database design
• CRUD operations

Frontend / Dashboard

• JavaScript
• HTML / CSS
• React or equivalent
• API integration
• WebSocket-based real-time updates
• Live video display

Deployment

• Docker / Docker Compose
• Linux / Ubuntu
• NVIDIA GPU / CUDA basics
• Environment configuration
• Logging and monitoring
• Basic networking

Edge AI

• Raspberry Pi
• NVIDIA Jetson Nano / Jetson platforms
• Edge AI model deployment
• Camera integration
• Real-time video processing
• Edge inference optimization
• IoT / sensor integration

Candidate Profile

Candidates should have hands-on project experience in Computer Vision and real-time video analytics and should be capable of taking a project from concept development testing deployment independently.

Candidates with only:

• Generic AI/ML experience
• Data Science experience
• Web Development experience
• Basic Python experience
• Theoretical Computer Vision knowledge

.

Preferred candidate profile

We are looking for someone with practical Computer Vision and real-time video analytics experience who can take a project from concept to working deployment. The developer should be able to learn and adopt different AI frameworks and libraries quickly. Experience with any suitable Computer Vision platform/framework such as YOLO, PyTorch, TensorFlow, Roboflow, Detectron2, MMDetection or equivalent is acceptable.