Venelin Valkov Courses

Nanonets OCR Small - Free Model to Turn Your Documents into Data for AI

Explore Nanonets OCR Small, a fine-tuned Qwen2.5-VL model that converts documents to Markdown, extracting tables, equations, signatures, and watermarks for AI applications.

• YouTube
• 18 minutes
• Self-Paced
• Free Video

The Illusion of Thinking - Can We Trust AI's Chain of Thought?

Uncover how AI models fabricate reasoning after choosing answers, creating plausible lies to hide their true decision-making process and why you shouldn't trust their explanations.

• YouTube
• 16 minutes
• Self-Paced
• Free Video

MedGemma 27B Local Multimodal Health AI Advisor - X-rays and Text-Only Diagnosis Test

Explore Google's MedGemma 27B multimodal AI model for healthcare diagnosis using X-rays and text analysis with real-world medical cases from Reddit's AskDocs community.

• YouTube
• 19 minutes
• Self-Paced
• Free Video

FLUX.1 Kontext [dev] Local Test - Image Generation and Edit with HuggingFace Open Weights Model

Explore FLUX.1 Kontext's 12B parameter model for AI image generation and editing with consistent character creation using HuggingFace's open weights implementation.

• YouTube
• 18 minutes
• Self-Paced
• Free Video

GLM-OCR (9B) - Local OCR Test - OCR, Document Extraction, Table Recognition

Discover GLM-OCR's 9B model capabilities for document OCR, extraction, and table recognition through hands-on testing and performance evaluation.

• YouTube
• 11 minutes
• Self-Paced
• Free Video

Is RAG Dead in 2026? - Build Local RAG from First Principles

Discover why RAG remains essential despite large context windows by building a local financial analyst agent from scratch using Python, Ollama, and TF-IDF retrieval.

• YouTube
• 14 minutes
• Self-Paced
• Free Video

Model Context Protocol (MCP) Tutorial - Connect Local AI Agent to Tools with MCP Server and Client

Learn how to implement Model Context Protocol (MCP) to connect local AI models to external tools. Build an MCP server, client, and Streamlit UI for a todo list manager using Ollama, Langchain, and AI agents.

• YouTube
• 38 minutes
• Self-Paced
• Free Video

Building a Local CAG System with Qwen3, Ollama and LangChain for Private Document AI Chatbots

Learn to build a fully local Cache-Augmented Generation (CAG) chatbot using Qwen3, Ollama, LangChain, and Streamlit that processes your private documents without external APIs or complex retrieval systems.

• YouTube
• 30 minutes
• Self-Paced
• Free Video

DeepSeek-R1 0528 for 100% Local Chat with Your Files - Financial Document Analysis AI with Ollama

Master private financial document analysis using DeepSeek-R1 locally with Ollama, comparing performance against cloud AI models through hands-on PDF processing.

• YouTube
• 25 minutes
• Self-Paced
• Free Video

Convert Any Document to LLM Knowledge with Docling and Ollama - 100% Local PDF to Markdown Pipeline

Master building a local PDF-to-Markdown pipeline using Docling & Ollama that preserves document structure, tables, and converts images to semantic descriptions for better RAG performance.

• YouTube
• 13 minutes
• Self-Paced
• Free Video

Gemini 3 Flash Tested - RAG, PDF Extraction, Resume Parser, Image Macronutrients Analysis

Discover Google's Gemini 3 Flash through real-world tests including RAG, PDF extraction, resume parsing, and image analysis to evaluate its production readiness.

• YouTube
• 16 minutes
• Self-Paced
• Free Video

MinerU 2.5 - Local OCR VLM Text and Table Extraction Test

Explore MinerU 2.5, a 1.2B vision-language model for two-stage OCR with text, table, and formula recognition capabilities compared to classical OCR approaches.

• YouTube
• 12 minutes
• Self-Paced
• Free Video

LangChain Quickstart with Local LLM - Ollama, Pydantic Structured Output, Tool Use, MLflow Tracing

Discover LangChain v1's streamlined API for swapping LLM providers, creating structured Pydantic outputs, and implementing custom tools with local Ollama setup in 15 minutes.

• YouTube
• 15 minutes
• Self-Paced
• Free Video

Embeddings and Vector Databases for Fast RAG - 100% Local with Ollama, Supabase, PostgreSQL and pgvector

Master semantic search for RAG systems using local embeddings, PostgreSQL, and pgvector to overcome keyword search limitations and build production-grade vector storage.

• YouTube
• 13 minutes
• Self-Paced
• Free Video

Advanced RAG Chunking - Contextual and Structural Chunking with LangChain and Ollama

Master advanced RAG chunking techniques using LangChain & Ollama to preserve document structure and inject contextual information, moving beyond simple character splitting methods.

• YouTube
• 14 minutes
• Self-Paced
• Free Video