AI/ML Developer & CV Intern at Wobot.ai
Building intelligent systems that see, learn, and adapt.
Passionate about computer vision, artificial intelligence, and solving real-world problems through AI. Currently contributing to AI-driven vision systems at Wobot.ai while exploring the frontiers of machine intelligence.
Who I Am
I'm Shreyansh Rao, an AI/ML developer with a deep interest in computer vision, machine learning systems, and building products that make a real impact. My journey in AI started with a curiosity about how machines can interpret visual information — and I've been building ever since.
Currently working as a CV Intern at Wobot.ai, I contribute to AI-driven vision analysis that helps businesses gain insights from camera feeds. My experience spans multiple internships across AI and ML domains, from NLP pipelines to real-time object detection systems.
When I'm not training models or debugging pipelines, I'm exploring new research papers, contributing to open-source projects, and experimenting with novel architectures on my personal projects.
Designing and training ML models for real-world applications
Object detection, segmentation, and visual understanding systems
PyTorch, TensorFlow, OpenCV, scikit-learn, and beyond
From idea to working prototype in record time
Where I Work
At Wobot.ai, I contribute to the development of AI-driven video intelligence systems that transform standard camera feeds into actionable business insights. My work involves building and optimizing computer vision pipelines from real-time object detection and tracking to scene understanding and behavioral analytics. I collaborate with the core engineering team to research, prototype, and deploy models that power Wobot's SaaS platform, helping enterprises automate visual inspection workflows and gain real-time operational intelligence.
My Journey
Working on real-time video analytics and computer vision pipelines. Building object detection and tracking models that power AI-driven workplace intelligence systems. Developing and fine-tuning deep learning models for scene understanding and behavioral analysis from CCTV feeds.
I deployed AI-powered applications to the cloud using Docker. I built a cloud-hosted platform for the ASB Audit Risk Analyzer, integrated an Ollama Phi-3 LLM for intelligent responses, and applied prompt engineering to improve output accuracy. This experience gave me strong exposure to production deployments, containerization, and scalable system design.
I focused on analyzing productivity bottlenecks in companies like ShareChat and CRED, researching their tech stacks, and recommending AI-powered automation opportunities within their development pipelines and also helped in developing Planto AI Copilot to automate software development workflows such as code generation, testing, and deployment.
What I've Built
A full-stack computer vision application that processes live video streams using YOLOv8, displaying real-time detection results in an interactive web dashboard. Supports multiple camera feeds and custom model fine-tuning.
Built a high-accuracy CNN-based image classifier trained on custom datasets. Implemented data augmentation pipelines, transfer learning from ResNet/EfficientNet, and deployed as a REST API with batch prediction support.
Developed a fine-tuned BERT-based sentiment analysis model for social media content. Includes a preprocessing pipeline for noisy text, handles multilingual inputs, and achieves 92%+ accuracy on benchmark datasets.
Built a real-time player tracking and re-identification system using YOLOv8 and DeepSORT. Mapped players across multiple video feeds using embeddings and cosine similarity to maintain consistent global IDs and improve tracking accuracy.
An intelligent chatbot that allows users to upload and interact with multiple PDFs simultaneously, enabling efficient information retrieval and contextual question answering using advanced NLP techniques. Supports multiple document uploads, semantic search, and context-aware responses.
A face recognition-based automated attendance system using dlib and face_recognition library. Real-time face encoding and matching, with a web interface for management and CSV export of attendance records.
Technical Stack
Academic Background
Specialization in Artificial Intelligence and Machine Learning. Coursework includes Deep Learning, Computer Vision, Data Structures & Algorithms, Probability & Statistics, Linear Algebra, and Software Engineering.
Recognition
Won in AI/ML hackathons by building innovative CV solutions under time constraints
Completed Andrew Ng's Deep Learning Specialization on Coursera with distinction
Certified TensorFlow Developer — proficiency in building and deploying ML models
Active contributor to ML/CV open-source repositories on GitHub
Co-authored a paper on computer vision techniques presented at a national conference
Recognized as a top-performing intern for contributions to CV pipeline efficiency
Get a full overview of my experience, skills, education, and projects in a clean, single-page format.
📄 Download PDFGet In Touch
I'm always open to discussing new opportunities, interesting AI/ML projects, or just having a conversation about the latest in computer vision and deep learning. Don't hesitate to reach out!