AI/ML Specialist | Open Source Contributor(GSSoC'26)

Hi,
I am Vidhi Gandhi !

Computer Science graduate (2026) specializing in AI/ML, with internship experience delivering production-ready ML systems, LLM-powered applications, and computer vision pipelines including a document analysis system with 94.35% pattern-detection accuracy.

View Projects Resume

Technical Skills

01

Languages & Core

Foundation in object-oriented programming and core concepts.

Python Java Data Structures & Algorithms OOP Deep Learning Computer Vision NLP Object Detection Statistical Modeling
02

AI & ML Frameworks

Designing intelligent pipelines, retrieval models, and autonomous agents.

PyTorch TensorFlow Scikit-learn HuggingFace Transformers OpenCV YOLO LangChain Ollama Google Gemini API RAG FAISS ChromaDB Prompt Engineering
03

Backend & Tools

Tools and platforms for deploying and managing robust codebases.

FastAPI Firebase Git/GitHub Jupyter Streamlit VS Code Anaconda

Projects

Featured architectures and models I have engineered.

ArcScanner Pro - PDF Pattern Detection

Demo
  • Engineered a web-based backend pipeline using PyMuPDF and OpenCV for PDF-to-image conversion, adaptive preprocessing, and multi-template matching with threshold tuning — achieving 94.35% average pattern localization accuracy.
  • Delivered real-time, high-resolution visualizations with bounding boxes and confidence scores, enabling instant architectural pattern analysis and download.
FastAPI • OpenCV • PyMuPDF • JavaScript

AI Smart Contract Auditor

Demo
  • Built a full-stack web application combining solc compiler validation with Gemini API analysis to scan Solidity smart contracts for security vulnerabilities, assign risk severity, and auto-generate corrected code.
  • Implemented JWT authentication, API-based contract fetching, and an interactive dashboard for real-time vulnerability visualization.
FastAPI • React • Solidity • Gemini API • SQLite • Python

Transformer-Based Collaborative Movie Recommendation System

GitHub | Demo
  • Designed and trained a sequential collaborative filtering model using a multi-head self-attention Transformer to learn user behavior patterns and predict personalized top-K movie recommendations.
  • Built a comprehensive data pipeline: dataset parsing, positive-rating filtering, movie tokenization, sequence generation, and fixed-length padding for efficient model training.
PyTorch • Streamlit • Python

Restaurant Review Analysis System (RAG)

GitHub | Demo
  • Developed a Retrieval-Augmented Generation (RAG) system using Ollama embeddings and ChromaDB semantic search to retrieve grounded context and generate evidence-based insights on food quality, pricing, and service.
  • Built an interactive Streamlit UI with cuisine filtering, restaurant selection, and natural-language query support for real-time review analytics.
LangChain • Ollama • ChromaDB • Streamlit

Professional Experience

AI/ML Engineer Intern

Jan 2026 - Present
Param Consulting Services, Inc. Vadodara
  • Engineered a document analysis system using PyMuPDF, OpenCV, and multi-template matching to detect input fields and structural PDF elements, achieving 94.35% average localization accuracy.
  • Built and fine-tuned ML/DL classification and detection models for image and document processing tasks, accelerating end-to-end analysis workflows.
  • Developed a Transformer-based Collaborative Movie Recommendation System (PyTorch) using multi-head self-attention to learn sequential user behavior patterns and generate top-K predictions.
  • Delivered a Smart Contract Auditor (FastAPI + Gemini API + Solidity compiler) that automates vulnerability detection, assigns severity levels, and generates corrected code — reducing manual audit time significantly.
  • Recognized as Best Sincere Intern for consistent performance, proactive ownership, and cross-functional contributions.

Data Analytics Intern

Aug 2023 - Sep 2023
YBI Foundation Remote
  • Designed and executed end-to-end data pipelines in Python and SQL, covering data ingestion, preprocessing, feature engineering, and ML model development.
  • Applied statistical modeling and advanced data structures to extract actionable business insights from complex, multi-source datasets, supporting data-driven decision-making.

Education & Achievements

Education

Drs. Kiran and Pallavi Patel Global University

2022 - 2026
B.Tech, Computer Science and Engineering | CGPA: 8.86 / 10 Vadodara

Certifications

  • Ethical Hacker — Cisco Networking Academy
  • Foundations of Cybersecurity — Google
  • Code Unnati Advanced 3.0 — Edunet Foundation
  • Python Bootcamp — Sanfoundry (Jun 2025)

Achievements & Open Source

  • Best Sincere Intern — Param Consulting Services, Inc. (2026)
  • GSSoC 2026 Contributor — AI Agents Track (active open-source contributor)
  • ODOO x MSU Hackathon 2025 Participant — MSU Baroda
  • AIU Anveshan 2024 — West Zone Presenter, DY Patil University