🔍 BugLens

BugLens is a production-ready Bug Similarity Search Engine designed to find duplicate or semantically similar bug reports in under 50ms. It leverages NLP tools (TF-IDF) and NearestNeighbors (via scikit-learn) built atop a lightning-fast asynchronous FastAPI web backend.

🌟 Highlights

• Intelligent Semantic Search: Understands and processes titles and descriptions to find contextually identical software bugs.
• Microsecond Inferences: Evaluates similarities between items with O(1) SQLite row mapping for instantaneous metadata retrieval.
• Zero Cold-Start: Employs FastAPI lifespan events to eager-load Machine Learning Models locally on application start.
• Scalable Architecture: Designed with future migrations in mind (FAISS, SBERT, Redis LRU Cache).
• Hardened Resiliency: Built-in defenses against DDoS payloads, integrated comprehensive PyTest validation, and formatted structured logging.

---

🏗️ Architecture

graph TD;
    Client-->|POST /search| FastAPI;
    FastAPI-->|Load on startup| Models[TF-IDF & NN Index];
    FastAPI-->|Transform & k-neighbors| Inquiry[Search Engine];
    Inquiry-->|Query Metadata| SQLite[(Bug Metadata DB)];
    SQLite-->FastAPI;
    FastAPI-->Client;

Loading

---

🛠️ Usage & Setup

1. Installation

Requires Python 3.11+.

# Clone the repository
git clone https://github.com/ShashankChinthirla/BugLens.git
cd BugLens/src

# Create and activate a virtual environment
python -m venv venv
source venv/bin/activate  # Or .\venv\Scripts\activate on Windows

# Install Dependencies
pip install -r requirements.txt

2. Training the Model

To initialize the models and populate the database from the data/raw/bugs.csv:

python training/train.py

This script versions models in models/v1/ and populates data/processed/metadata.db.

3. Running the API

Boot up the uvicorn ASGI server:

uvicorn app.main:app --reload

Navigate to http://127.0.0.1:8000/docs to use the auto-generated Swagger UI to test real-time search queries.

4. Running Tests

The repository includes unit integration tests and latency stress testers.

pytest tests/
python tests/evaluate.py
python tests/stress_test.py

---

📊 Benchmarks & Profiling

• Memory Footprint (TF-IDF CSR Matrix): Configured for max_features=5000. Memory peaks cleanly at ~100MB for 100k items.
• Precision Validation Structure: Our foundational static Precision@5 test successfully evaluates core accuracy thresholds over standard datasets.
• Logging & System State: Highly structured per-request latency tracking using Python extra dict layouts. Fully isolated try...finally database threads natively manage asynchronous DB connections.

🐳 Docker Containerization

BugLens is container-ready out of the box.

docker build -t buglens-api .
docker run -p 8000:8000 buglens-api