PopularMajor Project
Artificial Intelligence
Face Recognition Attendance System
Camera-based attendance with liveness detection and analytics
- Python
- OpenCV
- FaceNet
- Flask
- +2
₹4,900₹5,999
4.8/301 reviews
Artificial Intelligence · Capstone
AI Agent with Tool Calling
An LLM agent that plans steps and calls tools over structured data
4-5 weeks build
Overview
An agent that answers multi-step questions by planning and calling tools rather than guessing. Given a request about a sample sales database, it chooses between SQL queries, a calculator and a chart builder through an LLM API's function calling, observes each result and decides the next step. Every step is logged, and read-only database access keeps tools safe.
What makes it stand up
- 01Typed tool definitions with argument validation
- 02Plan, act and observe loop with a step limit
- 03Read-only SQL access with query allow-listing
- 04Full trace view of every tool call and result
- 05Task-completion evaluation on a written test suite
Module breakdown
- 01
Tools
SQL query, calculator and chart tools with schemas.
- 02
Agent loop
Function-calling loop with retries and a step budget.
- 03
Safety
Read-only roles, query checks and output limits.
- 04
Tracing
Stored traces of reasoning steps and tool outputs.
- 05
Interface
Next.js chat with inline tables, charts and traces.
After this, you will be able to
- Explain how function calling differs from plain prompting
- Defend the safety limits you placed on each tool
- Demo a multi-step question answered with its full trace
Technology stack
- Python
- LangChain
- FastAPI
- PostgreSQL
- Next.js
- Level
- Capstone
- Domain
- Artificial Intelligence
- Build time
- 4-5 weeks
- Difficulty
You receive
- Complete, commented source code
- 60-90 page project report (IEEE format)
- Editable presentation deck
- Architecture, ER and UML diagrams
- Local setup and deployment guide
- Sample dataset or seed data
- Viva question bank with answers
- One 45-minute walkthrough call
- 30 days of doubt-clearing support