Fine-Tuning & Model Customization Roadmap — Step-by-step Learning Path
Adapting foundation models to specific tasks and domains through fine-tuning, PEFT techniques and custom training data pipelines
The Fine-Tuning & Model Customization roadmap
- Fine-Tuning Fundamentals — beginner
- When to Fine-Tune vs Prompt vs RAG — beginner
- Fine-Tuning vs Prompt Engineering — beginner
- Fine-Tuning vs RAG — beginner
- Cost-Benefit Analysis of Fine-Tuning — beginner
- Common Fine-Tuning Use Cases — beginner
- Signs You Don't Actually Need Fine-Tuning — beginner
- Preparing Training Data for Fine-Tuning — beginner
- Building a Fine-Tuning Dataset — beginner
- Data Quality Over Data Quantity — beginner
- Labeling and Annotation for Fine-Tuning — beginner
- Data Augmentation Techniques — beginner
- Train/Validation/Test Splits for Fine-Tuning — beginner
- Training Fundamentals for Fine-Tuning — beginner
- Training Fundamentals for Language Models — beginner
- Learning Rates and Training Stability — beginner
- Batch Size and Gradient Accumulation — beginner
- Monitoring Training Runs — beginner
- Checkpointing During Training — beginner
- Parameter-Efficient Fine-Tuning (PEFT/LoRA) — beginner
- Parameter-Efficient Fine-Tuning (PEFT) Fundamentals — beginner
- LoRA: Low-Rank Adaptation — beginner
- QLoRA and Quantized Fine-Tuning — beginner
- Choosing LoRA Rank and Target Modules — beginner
- Merging and Managing LoRA Adapters — beginner
- Comparing PEFT Methods — beginner
- Full Fine-Tuning Approaches — beginner
- Full Fine-Tuning Fundamentals — beginner
- When Full Fine-Tuning Is Worth the Cost — beginner
- Distributed Training Basics — beginner
- Mixed Precision Training — beginner
- Continual Pre-Training — beginner
- Instruction Tuning and Chat Fine-Tuning — beginner
- Instruction Tuning Fundamentals — beginner
- Building Instruction-Following Datasets — beginner
- Chat Format Fine-Tuning — beginner
- System Prompt Consistency After Fine-Tuning — beginner
- Multi-Task Instruction Tuning — beginner
- RLHF and Preference-Based Alignment — beginner
- RLHF Fundamentals — beginner
- Reward Modeling — beginner
- Direct Preference Optimization (DPO) — beginner
- Collecting Human Preference Data — beginner
- Risks and Failure Modes of Alignment Training — beginner
- Evaluating Fine-Tuned Models — beginner
- Evaluating a Fine-Tuned Model — beginner
- Benchmark Selection for Fine-Tuned Models — beginner
- Comparing Base Model vs Fine-Tuned Model — beginner
- Human Evaluation of Fine-Tuned Output — beginner
- Detecting Regressions from Fine-Tuning — beginner
- Building a Fine-Tuning Evaluation Suite — beginner
- Hyperparameter Selection and Tuning — beginner
- Key Hyperparameters in Fine-Tuning — beginner
- Hyperparameter Search Strategies — beginner
- Early Stopping and Training Duration — beginner
- Learning Rate Schedules — beginner
- Reproducibility in Fine-Tuning Experiments — beginner
- Compute Infrastructure for Fine-Tuning — beginner
- GPU Requirements for Fine-Tuning — beginner
- Cloud vs On-Premise Training Infrastructure — beginner
- Using Managed Fine-Tuning Services — beginner
- Cost Estimation for Training Runs — beginner
- Managing Training Jobs and Experiment Tracking — beginner
- Overfitting and Catastrophic Forgetting — beginner
- Recognizing Overfitting in Fine-Tuned Models — beginner
- Catastrophic Forgetting — beginner
- Regularization Techniques — beginner
- Balancing Dataset Size and Diversity — beginner
- Mitigating Forgetting with Replay and Mixed Data — beginner
- Domain Adaptation Techniques — beginner
- Domain Adaptation Fundamentals — beginner
- Building Domain-Specific Vocabularies and Tokenization — beginner
- Fine-Tuning for Legal, Medical and Technical Domains — beginner
- Multilingual Fine-Tuning — beginner
- Style and Tone Adaptation — beginner
- Evaluating Domain Adaptation Success — beginner
- Deploying Fine-Tuned Models — beginner
- Deploying Fine-Tuned Models — beginner
- Serving Infrastructure for Custom Models — beginner
- A/B Testing Fine-Tuned vs Base Models — beginner
- Versioning Fine-Tuned Models — beginner
- Retraining and Model Refresh Cadence — beginner
- Quantization and Model Compression — beginner
- Quantization Fundamentals — beginner
- Post-Training Quantization — beginner
- Quantization-Aware Training — beginner
- Model Distillation — beginner
- Trade-offs Between Compression Techniques — beginner
- Licensing and Legal Considerations — beginner
- Model Licensing Fundamentals — beginner
- Commercial Use Restrictions on Open Models — beginner
- Data Licensing for Training — beginner
- Attribution and Derivative Model Requirements — beginner
- IP and Ownership Considerations for Fine-Tuned Models — beginner
- Career and Capstone — advanced