aishwaryanr
b939582394
* Restructure navigation into journeys and topics; backfill research and refresh courses - Add Use/Build/Understand journey pages and topic pages with a 101/201/301 catalog, surfaced through a Journey x Level grid in the README - Redistribute the free-course and notebook lists into the new navigation; add 2025-2026 courses and remove paid or dead entries - Backfill monthly best-papers lists from March 2025 through June 2026 - Extend the RAG, AI evaluation, and agentic search research tables to mid-2026 - Archive the 2024 course and paper material with banners - Fix the citation block and drop stale calls to action * Add all-free-courses-by-topic index; use numerals for numbers - Add courses.md: every free course grouped by topic, linked from the README router and the browse-by-topic section - Write counts as numerals across the navigation (10 not ten, 3-day not three-day) * Redesign the LLM foundations, agents, and RAG roadmaps for 2026 - Foundations roadmap: reorder so agents and evaluation are core days and fine-tuning becomes an optional advanced day; refresh all resources - Agents roadmap: rebuild around the model-plus-harness mental model, tools, context and memory, MCP and multi-agent, and agent evaluation - RAG roadmap: retrieval foundations, building a RAG app, agentic and advanced RAG - Point to current LevelUp materials and verified 2025-2026 resources * Refresh Agents 101 for 2026 and apply LevelUp Labs branding - Rewrite the Agents 101 guide around the model-plus-harness model, MCP, reasoning models, real-world agents (coding/computer-use/deep-research), and modern agent evaluation; drop the BabyAGI-era example and stale benchmarks - Rebrand references to LevelUp Labs and link the team to levelup-labs.ai * Archive the 2024 multimodal guide with a pointer to current material * Point each topic page to the canonical course list in courses.md * Add Harness Engineering path and enrich fine-tuning material - New Harness Engineering path: agent = model + harness, from using Claude Code, Cursor, and Codex to assembling and evaluating your own harness; wired into the Build journey's named paths - Fine-tuning topic: add current 2025-2026 material (post-training courses, the RLHF book, Hugging Face TRL and the LLM course chapter, Unsloth, Axolotl, and a practical 2025 fine-tuning guide) * Add role-based interview prep guide - New Role-Based Interview Prep: maps AI/LLM Engineer, ML/Fine-tuning Engineer, Applied Scientist, AI Product Manager, and Solutions Architect roles to what each interview tests and the repo material to prepare with; anchored on the 60 questions - Linked from the Interview Prep path and the README * Add branded roadmap headers and a real role-based interview question bank - Add LevelUp Labs branded headers for the LLM foundations and AI agents roadmaps (white background, blue, credited), wired into the guides - Rewrite role-based interview prep as an actual question bank with answers for AI/LLM Engineer, ML/Fine-tuning Engineer, Applied Scientist, PM, and Solutions Architect, rather than a navigation index * Add branded 3-Day RAG Roadmap header * Update RAG roadmap header with the branded 3-Day RAG image * Add deep role-based interview prep hub (AI Engineer, AI PM, FDE, AI Strategist) - Full folder per role: overview, interview rounds, a large answered question bank (210 questions across roles), verified free resources and courses, and a prep plan - Grounded in 2025-2026 research on how each role is actually interviewed at named companies; external links verified, cross-links to repo content - Replace the earlier shallow role summary with the hub and per-role folders
3.9 KiB
3.9 KiB
Topic: Fine-tuning
📚 Full course list: the complete, always-current set of free courses on this topic lives in All free courses: Fine-tuning and Post-training. The entries below are organized by journey and level.
Adapting a base model to your task: full fine-tuning, parameter-efficient methods (LoRA, PEFT), quantization, and RLHF.
Tags: Format 📖 Read / 🎥 Video / 🛠️ Notebook / 📝 Practice · Source ⭐ LevelUp Labs original / 🌐 External · year.
🏗️ Build 201
LevelUp Labs originals:
- Fine-tuning 101 ⭐ 📖 (2025): the basics of adapting a model, types of fine-tuning, and common challenges.
- Applied LLMs Mastery 2024, week 3: LLM fine-tuning ⭐ 📖 (2024) 🔴 archived.
Free external courses (full context on Build 201):
- Reinforcement Learning from Human Feedback 🌐 🎥 by DeepLearning.AI
- Finetuning Large Language Models 🌐 🎥 by DeepLearning.AI
- Open Source Models with Hugging Face 🌐 🎥 by DeepLearning.AI
- Quantization Fundamentals with Hugging Face 🌐 🎥 by DeepLearning.AI
- Quantization in Depth 🌐 🎥 by DeepLearning.AI
Hands-on notebooks:
- LLM Fine-tuning tutorials 🌐 🛠️ by ashishpatel26
- PEFT example notebooks 🌐 🛠️ by Huggingface
- Free LLM Fine-Tuning Notebooks 🌐 🛠️ by Youssef Hosni
🆕 Latest and hands-on (2025-2026):
- Post-training of LLMs 🌐 🎥 by DeepLearning.AI: turn a base model into an assistant.
- Reinforcement Fine-Tuning LLMs with GRPO 🌐 🎥 by DeepLearning.AI and Predibase.
- Fine-tuning and RL for LLMs: Intro to Post-training 🌐 🎥 by DeepLearning.AI and AMD.
- The RLHF and Post-Training Course and the free RLHF Book 🌐 📖 by Nathan Lambert: reward models, DPO, GRPO, RLVR.
- Fine-tuning chapter, LLM Course 🌐 📖 by Hugging Face.
- Fine-tune LLMs in 2025 and beyond 🌐 📖 by Philipp Schmid: a current, practical end-to-end guide.
- Unsloth 🌐 🛠️: fast, memory-efficient fine-tuning with free Colab notebooks.
- TRL, Transformer Reinforcement Learning 🌐 🛠️ by Hugging Face: the standard library for SFT, DPO, and GRPO.
- Axolotl 🌐 🛠️: a config-driven framework for fine-tuning open models.
🏗️ Build 301
- Training and Fine-Tuning LLMs for Production 🌐 📖 by Activeloop (production-grade fine-tuning).
Related topics: Foundations · Production and LLMOps. Journeys: Build. Back to the repository index.