A production-ready AI coding assistant built with DeepSeek Coder and Streamlit. This application provides intelligent programming help through a beautiful, responsive web interface with just pure coding expertise!
- 💬 Interactive Chat Interface with message history
- 🎨 Modern UI with gradient backgrounds and animations
- ⚡ Real-time Responses from DeepSeek Coder 1.3B model
- 📝 Code-Specific AI trained on 2 trillion tokens (87% code)
- 🔒 100% Local - Your code stays on your machine
- 🧠 Specialized for Programming - Answers only coding questions
graph TD
A[User Question] --> B[Streamlit UI]
B --> C[Ollama Local LLM]
C --> D[DeepSeek Coder 1.3B]
D --> E[Code Response]
E --> B
B --> F[(Chat History)]
| Layer | Technology | Purpose |
|---|---|---|
| Frontend | Streamlit | Beautiful web UI with zero HTML/CSS |
| LLM Engine | Ollama | Local model deployment & management |
| AI Model | DeepSeek Coder 1.3B | Specialized code generation & explanation |
| State Management | Streamlit Session | Chat persistence across interactions |
| Environment | Python venv | Clean dependency isolation |
# Example prompts you can ask:
"Explain async/await in Python with examples"
"Write a React component for a todo list"
"Debug this function: def add(a b) return a+b"
"What's the difference between let and var in JavaScript?"
"Create a REST API with Flask"- 🎨 Gradient backgrounds with smooth CSS animations
- 💭 Message bubbles with distinct user/bot styling
- ⏱️ Timestamps on every message for conversation tracking
- 🧹 One-click chat history clear button
- 📱 Responsive design that works on all devices
- ✨ Hover effects on all interactive elements
- ✅ Comprehensive error handling with graceful fallbacks
- ✅ Model availability checking before each interaction
- ✅ Session state management for persistent conversations
- ✅ Token usage tracking (estimated)
- ✅ Loading animations during response generation
- ✅ Temperature control for response creativity
| Skill | Implementation |
|---|---|
| LLM Integration | Connected DeepSeek Coder with Ollama for local deployment |
| Streamlit Development | Created professional web apps with pure Python |
| Ollama Management | Pulled, ran, and managed local LLM models |
| DeepSeek Coder | Leveraged specialized coding AI for programming tasks |
| Python venv | Isolated project dependencies cleanly |
| UI/UX Design | Custom CSS with gradients, animations, and responsive layouts |
| State Management | Session persistence across app reruns |
| Error Handling | Graceful degradation when models aren't available |
| Async Programming | Managed concurrent user requests |
- 📁 Project structure - Organized code for maintainability
- 🔧 Environment management - Virtual env best practices
- 📝 Documentation - Clear code comments and user guides
- 🐛 Debugging - Systematic problem-solving approach
- 🎨 UI Design - User-centered interface principles
- 🚀 Deployment - Local deployment strategies
# 1. Clone the repository
git clone https://github.com/yourusername/deepseek-coder-assistant.git
cd deepseek-coder-assistant
# 2. Create virtual environment
python -m venv .venv
# 3. Activate it (Windows)
.\.venv\Scripts\activate
# 4. Install dependencies
pip install -r requirements.txt
# 5. Pull the DeepSeek model
ollama pull deepseek-coder:1.3b
# 6. Verify the model is installed
ollama list
# 7. Run the app
streamlit run app.pystreamlit==1.54.0
ollama==0.6.1
pandas==2.3.3
numpy==2.4.2
pillow==12.1.1
python-dotenv==1.2.1🔐 You: Write a function to check if a string is a palindrome
🤖 Assistant: Here's a Python function to check palindromes:
```python
def is_palindrome(s):
"""
Check if a string is a palindrome.
Args:
s (str): Input string to check
Returns:
bool: True if palindrome, False otherwise
"""
# Remove non-alphanumeric characters and convert to lowercase
cleaned = ''.join(c.lower() for c in s if c.isalnum())
# Compare with its reverse
return cleaned == cleaned[::-1]
# Test cases
print(is_palindrome("A man, a plan, a canal: Panama")) # True
print(is_palindrome("race a car")) # False
print(is_palindrome("Was it a car or a cat I saw?")) # True
Time complexity: O(n) where n is the length of the string Space complexity: O(n) for the cleaned string
### **JavaScript Question**
🔐 You: Explain async/await with an example
🤖 Assistant: Async/await in JavaScript is syntactic sugar over Promises that makes asynchronous code look and behave more like synchronous code...
[Detailed explanation with multiple examples]
### **React Question**
🔐 You: Create a simple counter component in React
🤖 Assistant: Here's a basic counter component using React hooks:
import React, { useState } from 'react';
function Counter() {
const [count, setCount] = useState(0);
return (
<div style={{ textAlign: 'center', padding: '20px' }}>
<h2>Counter: {count}</h2>
<button onClick={() => setCount(count + 1)}>Increment</button>
<button onClick={() => setCount(count - 1)}>Decrement</button>
<button onClick={() => setCount(0)}>Reset</button>
</div>
);
}
export default Counter;
---
## 📈 **Performance Metrics**
| Metric | Value |
|--------|-------|
| **Model Size** | 776 MB (DeepSeek Coder 1.3B) |
| **Response Time** | 2-5 seconds average |
| **Context Window** | 4096 tokens |
| **Memory Usage** | ~1.5 GB RAM |
| **Languages Supported** | 80+ programming languages |
| **Training Data** | 2 trillion tokens (87% code) |
| **Accuracy** | 45.1% on HumanEval benchmark |
---
## 🚧 **Future Enhancements**
- [ ] **Multi-model Support** - Switch between different coding AIs
- [ ] **Code Execution** - Run generated code in a sandbox
- [ ] **Export Conversations** - Save chats to markdown files
- [ ] **Dark Mode** - Theme toggle for night coding
- [ ] **Code Highlighting** - Better syntax highlighting in responses
- [ ] **Voice Input** - Speech-to-text for hands-free coding
- [ ] **Favorite Prompts** - Save commonly used questions
- [ ] **Code Explanation Mode** - Toggle between brief/detailed explanations
---
**Built with ❤️ using Python, Streamlit, and DeepSeek Coder**
---
## ⚖️ **License**
MIT License - feel free to use this project for learning or commercial purposes!
---
## 🙏 **Acknowledgments**
- [DeepSeek AI](https://deepseek.com/) for releasing an amazing open-source coding model
- [Ollama](https://ollama.ai/) for making local LLM deployment effortless
- [Streamlit](https://streamlit.io/) for the incredible web framework that turns Python into UIs
- The open-source community for continuous inspiration and tools
---
## 📸 **Screenshots**
**

*Main chat interface with gradient background and message bubbles*
---
*Last Updated: February 17, 2026*
---
## ⭐ **Show Your Support**
If you found this project helpful, please give it a ⭐ on GitHub!
---
**Made with 🧠 + 💻 by [casss20](https://github.com/casss20/Deepseek_coder_assistant)**