Advanced GenAI & Safety

AI/ML Engineering Track | Phase 7

Best for: learners who want deeper adaptation, evaluation, alignment, and local fine-tuning workflows after they already understand the basics.

This phase is not the best first stop for new learners. It is where you go once model fundamentals and engineering discipline are already in place.

Modules

#	Module
1.1	Fine-tuning LLMs
1.2	LoRA & Parameter-Efficient Fine-tuning
1.3	Diffusion Models
1.4	RLHF & Alignment
1.5	Advanced Generation Techniques
1.6	LLM Evaluation
1.7	AI Red Teaming
1.8	AI Safety & Alignment
1.9	Modern PEFT: DoRA and PiSSA
1.10	Single-GPU Local Fine-Tuning
1.11	Multi-GPU and Home-Lab Fine-Tuning
1.12	Reasoning-Model RL: GRPO, RLVR, and DeepSeek-R1

Before You Start

You should already be comfortable with:

• generative AI fundamentals
• the difference between prompting, retrieval, and fine-tuning
• reproducible environments
• basic evaluation discipline

If not, go back to:

• Prerequisites
• Generative AI
• MLOps & LLMOps

Good Entry Points

• Fine-tuning LLMs for the overall decision framework
• Single-GPU Local Fine-Tuning if your actual path starts on one consumer GPU
• Modern PEFT: DoRA and PiSSA if you already know standard LoRA and want more modern adaptation techniques