Find a project you can actually build
Working IoT, AI/ML and embedded builds with abstracts, component lists, documentation, and developer support — ready for your final-year submission.
AI Agent for Automated Meeting Notes & Action Item Tracking
This project builds ai agent for automated meeting notes & action item tracking as an autonomous LLM-powered agent using a LangChain/LangGraph agent pipeline. The agent plans multi-step tasks, calls external tools/APIs, maintains memory/context, and reports outcomes back to the user with minimal manual intervention.
AI Agent for Automated Data Analysis & Report Generation
This project builds ai agent for automated data analysis & report generation as an autonomous LLM-powered agent using a ReAct-style tool-calling agent. The agent plans multi-step tasks, calls external tools/APIs, maintains memory/context, and reports outcomes back to the user with minimal manual intervention.
AI Coding Agent for Automated Unit Test Generation
This project builds ai coding agent for automated unit test generation as an autonomous LLM-powered agent using a Retrieval-Augmented Generation (RAG) agent. The agent plans multi-step tasks, calls external tools/APIs, maintains memory/context, and reports outcomes back to the user with minimal manual intervention.
AI Agent for Automated Resume Screening & Interview Scheduling
This project builds ai agent for automated resume screening & interview scheduling as an autonomous LLM-powered agent using a ReAct-style tool-calling agent. The agent plans multi-step tasks, calls external tools/APIs, maintains memory/context, and reports outcomes back to the user with minimal manual intervention.
AI Research Paper Agent for Literature Review Automation
This project builds ai research paper agent for literature review automation as an autonomous LLM-powered agent using a ReAct-style tool-calling agent. The agent plans multi-step tasks, calls external tools/APIs, maintains memory/context, and reports outcomes back to the user with minimal manual intervention.
Autonomous AI Research Assistant Agent with Web Search & Summarization
This project builds ai research assistant agent with web search & summarization as an autonomous LLM-powered agent using a LangChain/LangGraph agent pipeline. The agent plans multi-step tasks, calls external tools/APIs, maintains memory/context, and reports outcomes back to the user with minimal manual intervention.
Image-Based Food Calorie Estimation using Deep Learning
This project develops image-based food calorie estimation using a Convolutional Neural Network (CNN) trained on an appropriate image/video/audio/text dataset. The trained network is optimized for inference speed and served through a real-time demo application.
Deep Learning-Based Crop Disease Detection with CNN
This project develops crop disease detection with cnn using a 3D-CNN for spatio-temporal data trained on an appropriate image/video/audio/text dataset. The trained network is optimized for inference speed and served through a real-time demo application.
Human Activity Recognition using Wearable Sensor Data & DL
This project develops human activity recognition using a Transformer-based model (BERT/ViT) trained on an appropriate image/video/audio/text dataset. The trained network is optimized for inference speed and served through a real-time demo application.
Cell Image Classification for Cancer Detection using Deep Learning
This project develops cell image classification for cancer detection using a CNN-LSTM hybrid architecture trained on an appropriate image/video/audio/text dataset. The trained network is optimized for inference speed and served through a real-time demo application.
Video Summarization using Deep Learning
This project develops video summarization using a ResNet/EfficientNet backbone trained on an appropriate image/video/audio/text dataset. The trained network is optimized for inference speed and served through a real-time demo application.
Speaker Identification & Verification using Deep Learning
This project develops speaker identification & verification using a CNN-LSTM hybrid architecture trained on an appropriate image/video/audio/text dataset. The trained network is optimized for inference speed and served through a real-time demo application.