Back to Developers
Vansh Rajvanshi

Vansh Rajvanshi

Chennai 0+ yrs exp 67 · Good

About

No bio added yet

Skills & Expertise (48)

Deep Learning Intermediate
8.2/10
1
Years Exp
Computer Vision Intermediate
8.1/10
1
Years Exp
CNNs Intermediate
8.0/10
1
Years Exp
PyTorch Intermediate
7.6/10
1
Years Exp
TensorFlow Intermediate
7.6/10
1
Years Exp
Firebase localStorage MERN Stack Pandas NumPy Real-Time Systems Docker Docker Compose Ansible Git GitHub GStreamer Python C++ Java Arduino Random Forest Isolation Forest Plotly SciPy Chart.js JavaScript YOLO Deepsort OpenCV LLMs Model Optimization scikit-learn React Next.js Node.js Express SQLite Typescript HTML5 CSS3 Socket.IO WebSockets REST APIs Flask Streamlit MongoDb Postgresql

Work Experience

Deep Learning Research Intern

IntoAEC

Jun 2025 - May 2026

Built and deployed Estimate AI Agent (in production) — an agentic LLM-powered system that autonomously analyzes construction project requirements and generates full Bills of Quantities (BOQs) with real-time market pricing, replacing manual estimation workflows. Built and deployed Floor Plan Analyzer (in production) — a computer vision pipeline combining CNN-based semantic segmentation and one-shot learning to detect and classify architectural components (walls, rooms, doors, windows) and compute room-wise area for cost and design optimization. Implemented one-shot learning for architectural component detection, enabling the model to generalize to unseen floor plan layouts with minimal labeled data, improving coverage across diverse architectural styles. Engineered preprocessing pipelines and conducted multi-architecture benchmark comparisons (U-Net, DeepLab, custom CNNs) to maximize segmentation accuracy and robustness. Collaborated on end-to-end production deployment, managing model versioning, inference optimization, and integration with the core product platform.

AI Video Analytics Intern

Intel Unnati Industrial Training

Jun 2025 - Jun 2025

Architected real-time video analytics pipeline using DL Streamer and OpenVINO for intelligent media processing with object detection and classification on Intel hardware (Xeon CPU + iGPU). Developed Python automation framework to dynamically control GStreamer pipelines and orchestrate concurrent processing workflows, enabling scalable batch inference. Optimized deep learning model inference for Intel-specific hardware acceleration, improving throughput and processing accuracy through targeted model tuning.

Education

B.Tech — Computer Science & Engineering (Gaming Technology) - SRMIST

2023 - 2027 · Afghanistan

Certifications

Pipeline Creation using DL Streamer & System Scalability for Intel Hardware

Intel Unnati Industrial Training · 2025

Interested in this developer?

Profile Score Breakdown

📷 Photo 10/10
📄 Resume 10/10
💼 Job Title 0/10
✍️ Bio 0/10
🛠️ Skills 20/20
🎓 Education 10/10
⏱️ Experience 7/15
💰 Rate 0/5
🏆 Certs 5/5
Verified 5/5
Total Score 67/100

Profile Overview

Member sinceMay 2026

Availability Details

Visa Status

Citizen

Relocation

Open to Relocation