अधिकांश MCP सामग्री बड़े विचार पर ही रुक जाती है: AI टूल्स को बाहरी सिस्टम से जोड़ने का एक मानक तरीका। यह उपयोगी है, लेकिन जब आप किसी Python प्रोजेक्ट में बैठे यह सोच रहे हों कि पहले क्या बनाना है,...
Khushal Jethava
AI & Python Tutorials by Khushal Jethava
Practical guides on LLMs, RAG, MLOps, computer vision, and Python programming from a Machine Learning Engineer working with Generative AI. Covering fine-tuning, AI agents, explainable AI, and production ML pipelines with hands-on code examples.
Build a Reasoning LLM from Scratch in Python
Build a reasoning LLM from scratch in Python — a BPE tokenizer, RoPE attention, SwiGLU transformer blocks, and chain-of-thought dual-loss training. No APIs, no wrappers, just pure PyTorch.
Understanding Embeddings: From Word2Vec to Modern LLMs
Learn how embeddings work in Python, from Word2Vec to modern transformer-based embedding models. Covers vector arithmetic, cosine similarity, and visualizing embeddings with t-SNE.
Feature Engineering for Tabular ML: A Python Guide
Learn practical feature engineering techniques in Python for tabular machine learning — encoding, scaling, binning, interaction features, and target encoding with real examples.
Building a Local AI Chatbot with Ollama and Python
Learn how to build a private, local AI chatbot in Python using Ollama. Covers installation, streaming responses, conversation memory, and a simple Gradio web UI.
Hyperparameter Tuning with Optuna: A Python Guide
Learn hyperparameter tuning in Python with Optuna. Covers search spaces, pruning, multi-objective optimization, and a complete example tuning an XGBoost model.
Prompt Caching Explained: How It Cuts LLM Costs
Learn how prompt caching works with Claude and OpenAI APIs in Python. See real cost and latency comparisons, and learn how to structure prompts to maximize cache hits.
Building a RAG Evaluation Pipeline with Python
Learn how to evaluate RAG systems in Python using retrieval metrics, faithfulness scoring, and RAGAS. Build a complete evaluation pipeline to catch hallucinations before production.
Python Async/Await for AI Pipelines: A Practical Guide
Learn how to use Python async/await to speed up LLM API calls, batch embeddings, and build concurrent AI pipelines with asyncio and httpx. Includes real benchmarks.
Quantization for LLMs: Run Big Models on Small Hardware
Learn how LLM quantization works in Python — GPTQ, GGUF, and bitsandbytes 4-bit/8-bit. See real memory and speed comparisons and run a quantized model on a laptop GPU.