Descriptions
Ollama & Local LLMs: Fine-Tune, Deploy, Build Python AI Apps, If you are tired of black-box cloud APIs and want AI that runs on your hardware, keeps your data local, and still ships like a real product, this course gives you a complete path from your first Python call to a deployable model and a capstone app you can show in a portfolio. You start with Ollama and Python (chat, streaming, and generation patterns you can reuse everywhere), then move into Streamlit apps, RAG-style assistants, multi-agent workflows, Unsloth + QLoRA fine-tuning, and exporting models to Ollama. You finish with a full local coding assistant featuring a browser, editor, diffs, code and shell execution, and optional web research and vision hooks. Theory is always connected to working code and a clear next step. You build rather than only watch: practical projects include chat UIs, PDF Q&A, embedding search, personal note and diary apps, CrewAI agent teams, Whisper + Ollama video Q&A, and a large Streamlit coding IDE powered by Ollama. You also fine-tune a small instruction model, compare base and fine-tuned answers, merge adapters, and ship the result through Ollama, with clear explanations of quantization and model quality so your exports behave as expected.
Throughout the course, you connect Ollama to Python using chat, streaming, and generation patterns; build Streamlit front ends on local models, including vision demos and media workflows with video frames, Whisper transcription, and questions about audio; and create retrieval-based assistants using PDF context, LangChain templates and chains, long-text chunking, and spaCy embedding search from similarity through full chunk retrieval. You orchestrate multi-step AI with LangChain and CrewAI, run Hugging Face models locally, and explore text-to-speech and text-to-video when your hardware allows. You train a personalized small LLM with Unsloth + QLoRA, monitor validation for catastrophic forgetting, compare it with the base model, merge adapters, and package it for Ollama using a Modelfile and practical export notes. The capstone is a local LLM-assisted coding environment with a project browser, editor, streaming chat, diff-reviewed changes, code and shell execution, optional web research, planning, persistence, and more. Local LLMs give you privacy, predictable costs, and the freedom to specialize models on your data and ship prototypes without relying on a single cloud provider.
What you’ll learn
- Use Ollama from Python for chat, streaming, and generation.
- Build Streamlit user interfaces on top of local models.
- Add vision capabilities by using images with multimodal models, such as LLaVA-style workflows.
- Process video frames, transcribe audio with Whisper, and answer questions about audio.
- Build PDF question-and-answer workflows and prompts grounded in documents.
- Use LangChain with Ollama integration, templates, LLMChain, and prompts with multiple inputs.
- Split long text into chunks and summarize it with map-style workflows.
- Use spaCy embeddings for similarity, semantic search, and RAG over long text.
- Build personal knowledge apps for notes and diary entries.
- Run CrewAI agents with Ollama, including sequential workflows and domain-specific examples.
- Run Hugging Face Transformers locally and manage cache and disk usage.
- Try text-to-speech and text-to-video scripts when the GPU supports them.
- Run Unsloth inference, for example with SmolLM2, and inspect tokenizer and model behavior.
- Fine-tune models with Unsloth + QLoRA and monitor validation for forgetting.
- Compare base and fine-tuned models and merge LoRA adapters.
- Export models to Ollama using a Modelfile while considering quantization trade-offs.
- Complete a capstone local LLM coding assistant with an editor, chat, diffs, code and shell execution, optional web research, planning, and persistence.
Who this course is for
- Python developers who want to add local LLMs to their projects.
- Beginner-to-intermediate coders who are comfortable installing packages and running scripts.
- Builders who want Streamlit apps powered by Ollama, not just command-line chat.
- Privacy-minded learners who prefer on-device AI and local data.
- Hobbyists and professionals exploring RAG, LangChain, embeddings, and agents without relying only on the cloud.
- Creators who want multimedia workflows involving images, video, and Whisper audio.
- Practitioners ready to fine-tune a small model with Unsloth and QLoRA and deploy it with Ollama.
- Portfolio-focused students who want a capstone project, such as a local coding-assistant app.
- Windows, Linux, or macOS users willing to troubleshoot GPU, Ollama, and Python environments when needed.
Specificatoin of Ollama & Local LLMs: Fine-Tune, Deploy, Build Python AI Apps
- Publisher : Udemy
- Teacher : Abdurrahman TEKIN
- Language : English
- Level : All Levels
- Duration : 16 hours and 38 minutes
Content of Ollama & Local LLMs: Fine-Tune, Deploy, Build Python AI Apps
Requirements
- Basic Knowledge of Python
Pictures
Sample Clip
Installation Guide
Extract the files and watch with your favorite player
Subtitle : English
Quality: 1080p
Download Links
Password file(s): www.downloadly.ir
File size
22.31 GB