The ultimate curated collection of AI tools, frameworks, datasets, papers, courses and communities — everything you need to build, learn, and scale AI in one place.
From LLMs to computer vision, MLOps to AI ethics, this repo is your go-to hub for high-quality, up-to-date AI resources.
| You Want | We Deliver |
|---|---|
| One source of truth for AI tools | 500+ vetted links |
| Beginner-friendly learning paths | Curated courses & tutorials |
| Production-ready MLOps | Tools, pipelines, monitoring |
| Cutting-edge research | Papers, models, benchmarks |
| Active community | Discord, forums, events |
- Learning & Education
- Frameworks & Libraries
- MLOps & Deployment
- Datasets
- AI Tools & Apps
- Ethics, Safety & Governance
- Communities & Events
- Trending AI Tools (2025-2026)
- Hands-On Quick Start Guides
- Tool Comparisons
- Real-World Use Cases
- AI Learning Roadmap
- Expert Tips & Best Practices
- AI Interview Preparation
- Hidden Gems
- Contribute
- CS229: Machine Learning (Stanford)
- Deep Learning Specialization (Coursera)
- Fast.ai Practical Deep Learning
- Hugging Face NLP Course
- PyTorch – Dynamic & research-friendly
- TensorFlow – Production-scale
- JAX – High-performance ML
- LangChain – Build LLM apps
- LlamaIndex – RAG & data connectors
- Hugging Face Transformers
- Cursor – AI-first code editor with Claude integration
- GitHub Copilot – AI pair programmer by GitHub & OpenAI
- Replit AI – Code, deploy, and collaborate with AI
- v0 by Vercel – Generate UI components from text prompts
- bolt.new – Full-stack web app development with AI
- Codeium – Free AI code completion and chat
- Tabnine – AI assistant for code completion
- Windsurf – Agentic IDE by Codeium
- Claude Code – Command-line AI coding agent
- ZeroGTM – Open-source AI SDR agent for automated outbound sales prospecting and outreach
What's Hot Right Now — The latest AI tools making waves in the community
- OpenAI o1 – Next-gen reasoning model (Dec 2024)
- Google Gemini 2.0 – Multimodal powerhouse
- Anthropic Claude 3.5 Sonnet – Best coding assistant (200K context)
- Perplexity Pro – AI-powered search engine
- NotebookLM – Google's AI note-taking with audio insights
- Sora by OpenAI – Text-to-video generation (limited access)
- Pika Labs – Video generation from text/images
- HeyGen – AI avatar & video translation
- Zed AI – Lightning-fast multiplayer code editor with AI
- Pieces for Developers – AI-powered code snippets manager
- Phind – AI search engine for developers
- Continue.dev – Open-source Copilot alternative
- Cody by Sourcegraph – AI coding assistant with codebase context
- Magnific AI – AI image upscaler (insane quality)
- Leonardo.AI – Game asset generation
- Ideogram – Text rendering in images (finally!)
- CapCut AI – AI video editing for creators
# pip install langchain openai
from langchain.chat_models import ChatOpenAI
from langchain.schema import HumanMessage
llm = ChatOpenAI(model="gpt-4")
response = llm([HumanMessage(content="Explain AI in 3 sentences")])
print(response.content)# pip install transformers pillow
from transformers import pipeline
classifier = pipeline("image-classification")
result = classifier("path/to/image.jpg")
print(result)# pip install llama-index
from llama_index import VectorStoreIndex, SimpleDirectoryReader
documents = SimpleDirectoryReader('data').load_data()
index = VectorStoreIndex.from_documents(documents)
query_engine = index.as_query_engine()
response = query_engine.query("Your question here")
print(response)Week 1: Foundations
- Day 1-2: Complete Fast.ai Lesson 1
- Day 3-4: Build a simple classifier
- Day 5-7: Fine-tune a pre-trained model
Week 2: Deep Learning
- Day 8-10: PyTorch basics
- Day 11-14: CNN for image recognition
Week 3: LLMs & NLP
- Day 15-17: Hugging Face Transformers
- Day 18-21: Build a RAG chatbot
Week 4: Production
- Day 22-25: Deploy with FastAPI
- Day 26-28: Add monitoring
- Day 29-30: Portfolio project showcase
| Feature | OpenAI GPT-4 | Claude 3.5 | Gemini 2.0 | Open Source (Llama 3) |
|---|---|---|---|---|
| Best For | General use | Coding | Multimodal | Privacy/Cost |
| Context | 128K | 200K | 2M | 8K-128K |
| Pricing | $$ | $$ | $ | Free |
| Speed | Fast | Fast | Very Fast | Varies |
| Code Quality | ⭐⭐⭐⭐ | ⭐⭐⭐⭐⭐ | ⭐⭐⭐⭐ | ⭐⭐⭐ |
| Reasoning | ⭐⭐⭐⭐⭐ | ⭐⭐⭐⭐ | ⭐⭐⭐⭐ | ⭐⭐⭐ |
| Database | Best For | Speed | Scalability | Ease of Use |
|---|---|---|---|---|
| Pinecone | Production | ⭐⭐⭐⭐⭐ | ⭐⭐⭐⭐⭐ | ⭐⭐⭐⭐⭐ |
| Weaviate | Hybrid search | ⭐⭐⭐⭐ | ⭐⭐⭐⭐⭐ | ⭐⭐⭐⭐ |
| Qdrant | Open-source | ⭐⭐⭐⭐⭐ | ⭐⭐⭐⭐ | ⭐⭐⭐⭐ |
| ChromaDB | Local dev | ⭐⭐⭐⭐ | ⭐⭐⭐ | ⭐⭐⭐⭐⭐ |
| FAISS | Research | ⭐⭐⭐⭐⭐ | ⭐⭐⭐ | ⭐⭐⭐ |
-
Customer Support Automation
- Stack: LangChain + GPT-4 + Pinecone
- ROI: 60% reduction in support tickets
- Example Repo
-
Code Review Assistant
- Stack: Claude 3.5 + GitHub API
- Impact: 40% faster PR reviews
- Tutorial
-
Document Intelligence
- Stack: Azure Document Intelligence + OpenAI
- Use: Extract data from invoices, contracts
- Demo
- AI Tutor: Personalized learning paths using RAG
- Essay Grader: Automated feedback with rubrics
- Language Learning: Conversation practice bot
- AI Game NPCs: Dynamic dialogue with memory
- Music Generator: Style transfer for audio
- Story Writer: Interactive fiction with AI
graph LR
A[🌱 Beginner] --> B[📊 Python + Math]
B --> C[🤖 ML Basics]
C --> D[🧠 Deep Learning]
D --> E[🔮 LLMs & Transformers]
E --> F[🚀 Production & MLOps]
F --> G[⭐ AI Expert]
- ✅ Python fundamentals
- ✅ NumPy, Pandas basics
- ✅ Linear algebra & calculus refresher
- ✅ Statistics fundamentals
- Project: Predict house prices with linear regression
- ✅ Scikit-learn
- ✅ Classification, regression, clustering
- ✅ Model evaluation & cross-validation
- ✅ Feature engineering
- Project: Image classifier (cats vs dogs)
- ✅ PyTorch or TensorFlow
- ✅ CNNs, RNNs, Transformers
- ✅ Transfer learning
- ✅ Hyperparameter tuning
- Project: Object detection with YOLOv8
- ✅ Hugging Face ecosystem
- ✅ Prompt engineering
- ✅ Fine-tuning LLMs
- ✅ RAG systems
- Project: Build a domain-specific chatbot
- ✅ Docker & Kubernetes
- ✅ FastAPI / Flask
- ✅ Model monitoring
- ✅ CI/CD for ML
- Project: Deploy scalable API
Total Time: 10-15 months to job-ready
- Be Specific: ❌ "Write code" → ✅ "Write a Python function that validates email addresses using regex"
- Use Examples: Few-shot learning dramatically improves output
- Set Context: "You are a senior Python developer..."
- Iterate: Start broad, then refine
- Chain of Thought: Ask the model to "think step by step"
- Cache embeddings: Don't re-embed the same content
- Use smaller models: GPT-3.5 is 10x cheaper than GPT-4
- Streaming: Better UX and you can stop early
- Batch requests: Save 50% on API costs
- Monitor usage: Set up alerts for spend
- Never put API keys in code (use env variables)
- Implement rate limiting
- Sanitize user inputs
- Use prompt injection防护
- Audit AI outputs before production use
- Async everything: Use
asynciofor I/O operations - Vector DB indexing: HNSW > IVF for most use cases
- Quantization: Run LLMs 4x faster with minimal quality loss
- Caching: Redis for frequent queries
Technical:
- Explain transformers architecture
- Difference between RAG and fine-tuning?
- How do you handle hallucinations?
- Explain attention mechanism
- What is RLHF?
System Design:
- Design a recommendation system
- Build a scalable chatbot architecture
- Create a real-time fraud detection system
🌟 Hidden Gems: Underrated Tools
- Marvin – Build AI-powered software
- Instructor – Structured outputs from LLMs
- DSPy – Programming foundation models
- Autogen – Multi-agent conversations
- LiteLLM – Universal LLM API
- Mem0 – Memory layer for AI apps
- Lamini – Enterprise LLM platform
We welcome contributions!
Found a missing gem? Want to add a new category?
- Fork the repo
- Add your resource in the correct section
- Follow format:
- [Name](link) – Short description. - Open a PR!
Note
This repository is automatically synced with our Interactive Web Portal. Any changes merged into the main branch will reflect on the website within minutes!
Developed with ❤️ by Dhanush Nehru
