DeepCode

AI-powered code understanding platform for developers and learners.

Live at → adarshsalukhe.github.io/DeepCode/

---

What It Does

Paste any code and get instant AI-powered analysis — powered by GPT-OSS 120B via Groq.

Mode	Description
Understand	Line-by-line explanation at beginner, intermediate, or expert level
Debug	Root cause analysis with mechanical explanation and fix
Complexity	Time and space complexity with Big O breakdown
Bug Finder	Proactive bug detection — no error message needed
Challenge	Mixed quiz — MCQ, complete the function, write a test

---

Supported Languages

Python · JavaScript · TypeScript · Rust · Go · C++ · C · Ruby · HTML · SQL

---

Features

• 🔐 Google and GitHub OAuth via Supabase
• ⚡ Streaming responses — results appear token by token
• 📊 Usage tracking — see who used what in Supabase dashboard
• 🕐 History panel — revisit your last 20 analyses
• 🛡️ Rate limiting, input validation, CORS protection
• 📏 Explanation length control — Short / Medium / Detailed

---

Tech Stack

Layer	Technology
LLM	GPT-OSS 120B via Groq API
Backend	FastAPI + Python
Auth & DB	Supabase
Frontend	Vanilla React (single HTML file)
Streaming	Server-Sent Events (SSE)
Hosting	Railway (backend) + GitHub Pages (frontend)

---

Local Development

Prerequisites

• Python 3.11+
• Groq API key — console.groq.com
• Supabase project — supabase.com

Backend

cd backend
pip install -r requirements.txt

Create backend/.env:

LLM_PROVIDER=groq
GROQ_API_KEY=your_groq_api_key

SUPABASE_URL=https://your-project.supabase.co
SUPABASE_SERVICE_KEY=your_service_role_key

ALLOWED_ORIGINS=*
RATE_LIMIT_REQUESTS=10
RATE_LIMIT_WINDOW=60

Start backend:

python -m uvicorn main:app --reload --port 8000

Frontend

cd frontend/public
python -m http.server 3000

Open http://localhost:3000

---

Supabase Setup

Run this SQL in your Supabase SQL editor:

create table usage_logs (
  id          uuid default gen_random_uuid() primary key,
  user_id     uuid references auth.users(id),
  user_email  text,
  mode        text,
  language    text,
  level       text,
  code_length int,
  created_at  timestamptz default now()
);

create table user_logins (
  id           uuid default gen_random_uuid() primary key,
  user_id      uuid references auth.users(id),
  user_email   text,
  provider     text,
  logged_in_at timestamptz default now()
);

alter table usage_logs enable row level security;
alter table user_logins enable row level security;

create policy "insert own usage" on usage_logs
  for insert with check (auth.uid() = user_id);

create policy "insert own login" on user_logins
  for insert with check (auth.uid() = user_id);

---

Project Structure

DeepCode/
├── backend/
│   ├── main.py                  # FastAPI app + security middleware
│   ├── requirements.txt
│   ├── routes/
│   │   ├── explain.py           # Understand mode
│   │   ├── debug.py             # Debug mode
│   │   ├── challenge.py         # Challenge mode
│   │   └── analyze.py           # Complexity + Bug Finder
│   └── services/
│       ├── llm.py               # Multi-provider LLM client
│       ├── prompts.py           # All prompt templates
│       ├── parser.py            # AST code parser
│       ├── sandbox.py           # Python code executor
│       ├── stream_utils.py      # SSE streaming utilities
│       └── auth.py              # Supabase JWT verification
└── frontend/
    └── public/
        └── index.html           # Entire frontend — single file

---

Security

• JWT verification on every API route via Supabase
• Rate limiting — 10 requests per minute per IP
• Input validation — code length, language whitelist, field checks
• CORS locked to frontend domain in production
• API docs disabled in production
• Secrets never in frontend code

---

Built By

Adarsh Salukhe — AI/LLM Engineer

GitHub · LinkedIn