IIT Madras graduate building production-ready AI products with Python, LLMs, RAG, and computer vision.

Abhishek Jha  • Peerlist

Built a decentralized trust infrastructure for autonomous AI agents featuring verifiable identities, reputation management, governance mechanisms, and blockchain-backed audit trails. Integrated AI capabilities with decentralized technologies to improve transparency, accountability, and trust in multi-agent ecosystems.

AgentTrust – Decentralized Trust Layer for AI

Created an autonomous multi-agent AI platform capable of generating branching cinematic storylines using coordinated AI agents. Leveraged LangGraph, Qwen, and Alibaba Cloud services to orchestrate story generation, scene planning, and AI-driven creative workflows for dynamic interactive filmmaking.

Butterfly Director AI – Autonomous AI Film St

Developed end-to-end customer churn prediction systems using Artificial Neural Networks

Customer Churn Prediction (ANN) - Deep Learni

Built and trained a deep Convolutional Neural Network (CNN) to classify dog and cat images using the Kaggle dataset.

Dogs vs Cats Image Classification (CNN) - Dee

Developing a scalable full-stack operations platform for managing sound and light rental businesses. • Architected a monorepo-based system managing 18+ interconnected database tables and 12+ type-safe API routes using modern backend tools • Implemented a secure Role-Based Access Control (RBAC) system with Admin, Manager, and Staff roles for structured workflow management • Led backend development in a 4-member agile team, conducting code reviews and managing sprint planning using Scrum methodologies • Built a high-performance frontend using React and Tailwind, ensuring a responsive and intuitive user experience • Integrated AI-assisted development workflows to accelerate feature development and automate testing of critical business logic Focused on scalability, performance, and real-world usability for small business operations.

EventFlow – Scalable Operations Platform for 

Developed an AI-powered healthcare facility verification system that evaluates hospital capabilities across critical care services using explainable AI and evidence-based trust scoring. The platform assists users in assessing healthcare facilities through transparent AI-driven recommendations.

Facility Trust Desk – AI Healthcare Verificat

Automatically creates comprehensive notes and quiz questions from any given topic

Learn and Test Yourself - Gen AI

Explored the dataset of 116K+ rows and 52 features, performing extensive EDA (because honestly, EDA can make or break your model!). Tackled challenges like missing values, skewed data, and high-cardinality categorical features. Engineered new features such as user_avg_purchase_value, month, dayofweek, weekday, weekend which turned out to be super important (50%+ importance for user_avg_purchase_value). Tried multiple ML models – Random Forest, Linear Regression, XGBoost – with XGBoost giving me the best accuracy of 70% on test data. Learned how crucial feature engineering + data preprocessing are in driving performance.Explored the dataset Among Top 3% of the leaderboard (55/1700+)

Machine Learning Regression Project - Engage2

Developed a production-style full-stack platform that optimizes cross-border remittances by comparing stablecoin, Ethereum Layer-2, and traditional transfer routes based on transaction cost, speed, risk, and recipient value. Built an AI-powered recommendation engine that provides transparent route analysis, natural language explanations, and transfer optimization. Designed the application using a modern microservice architecture with a Next.js frontend, FastAPI backend, Solidity smart contracts, Docker-based deployment, and automated CI/CD workflows.

RemitAI – AI-Powered Cross-Border Remittance 

Sentiment analysis and response from user reviews using advanced NLP techniques.

Sentiment Responder - NLP

Designed and developed an AI-powered collaborative productivity platform that unifies workspace management, Kanban boards, calendars, notes, whiteboards, voice commands, and intelligent AI assistants into a single application. Implemented real-time collaboration features, intelligent workflow automation, and AI-powered productivity tools to enhance team communication and project management.