IntelliReview AI

Home Page
Result Page
Repository Summary
Dependency Graph
Full Report
History and Download Full report

Inspiration

What it does

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned

What's next for IntelliReview AI

Inspiration

We both work as undergraduate students on the same engineering team, and after spending over a year working on large codebases, we realized how difficult it was to understand code written by others and the exact purpose behind every module.

Whenever we found bugs, code smells, or architectural issues, we would manually note them down on paper to keep track of them during development. This process was repetitive, time-consuming, difficult to maintain, and inefficient as projects continued to grow.

At the same time, we looked at how modern software engineering is evolving. Today's software systems are becoming increasingly large and complex, making manual code reviews difficult to scale. Traditional static analysis tools can detect issues, but they usually present isolated warnings without explaining their overall impact on the health and maintainability of an entire project.

We wanted a solution that could analyze an entire repository, identify the most important engineering problems, prioritize them based on their impact, explain why they matter, and generate actionable insights in one place.

That idea became IntelliReview AI-an AI-powered Engineering Intelligence Platform that combines repository analysis, dependency visualization, technical debt estimation, architecture evaluation, security analysis, and AI-generated recommendations to help developers understand, review, and improve their codebases more effectively.

What it does

IntelliReview AI is an AI-powered Engineering Intelligence Platform that analyzes source code at both file and repository levels.

The platform currently supports:

Single file code review
Repository ZIP analysis
GitHub repository analysis
Pull Request review
AI-generated code review reports
Repository health scoring
Technical debt analysis
Interactive dependency graph visualization
Circular dependency detection
Complexity heatmap generation
Module risk ranking
Security vulnerability detection
Root cause analysis
Repository architecture analysis
AI-powered fix suggestions
Historical quality comparison
Executive summaries
PDF and text report generation

The goal is to help developers quickly understand the health of an entire software project rather than manually reviewing thousands of lines of code.

How we built it

The application is built using a modular architecture where every analysis component is implemented as an independent analyzer.

The backend performs:

Repository extraction and parsing
AST-based code analysis
Static analysis
Repository metrics collection
Dependency graph construction
Technical debt scoring
Security analysis
Architecture evaluation
Risk ranking
Executive summary generation

Google Gemini powers the AI review engine, generating human-readable insights and actionable recommendations from the collected analysis.

The frontend is implemented using Streamlit, providing an interactive dashboard with visualizations, metrics, and downloadable reports.

Challenges we ran into

One of the biggest challenges was combining multiple analysis techniques into a single coherent workflow.

Some of the engineering challenges included:

Building repository-wide dependency graphs
Detecting circular dependencies
Calculating repository-level technical debt
Ranking modules based on engineering risk
Integrating AI-generated insights with deterministic static analysis
Keeping the architecture modular while adding new analysis capabilities

Balancing detailed analysis with usability was another major challenge, as the platform needed to present complex engineering information in a way that developers could quickly understand.

Accomplishments that we're proud of

We successfully built a platform capable of analyzing complete repositories instead of individual files.

Some highlights include:

Interactive dependency visualization
Repository health scoring
Technical debt estimation
Complexity heatmaps
Architecture analysis
Security analysis
AI-generated engineering recommendations
Professional PDF reporting
Support for GitHub repositories and ZIP uploads
Modular analyzer architecture that can be extended with additional analysis engines

What we learned

Developing IntelliReview AI demonstrated that combining deterministic static analysis with large language models provides significantly more actionable insights than either approach alone.

We also learned the importance of modular software architecture when building engineering tools that continue to evolve with new analysis capabilities.