Lessons
view all ⭢
lesson
DeepSeek-V3 ArchitectureAI BootcampExplore DeepSeek-V3’s MLA, MoE, MTP, FP8 training, and understand transformer optimizations.
lesson
Positional Encoding in TransformersAI BootcampUnderstand positional encoding (sinusoidal, RoPE, learned), compare types, and study skip connections/layer norms.
lesson
Text-to-Voice PipelinesAI BootcampInvestigate pipelines from text prompts through intermediate representations to audio. Compare zero-shot vs fine-tuned models in speech generation.
lesson
Text-to-SQL SystemsAI BootcampImplement text-to-SQL with structured prompts, train/evaluate SQL generation accuracy using execution-based metrics.
lesson
Agent Architectures and ToolkitsAI BootcampCompare single-agent vs multi-agent architectures, explore AutoGen, LangGraph, CrewAI use cases.
lesson
Agent Design PatternsAI Bootcamp- Understand agent design patterns: Tool use, Planning, Reflection, Collaboration - Learn evaluation challenges in agent systems: output variability, partial correctness - Study architecture patterns: single-agent vs constellation/multi-agent - Explore memory models, tool integration, and production constraints - Compare agent toolkits: AutoGen, LangGraph, CrewAI, and practical use cases
lesson
Designing AI Code AgentsAI BootcampDesign a multi-agent AI IDE stack with chunking, AST parsing, and RAG+LLM collaboration.
lesson
Reverse Engineering Vibe Coding AgentsAI BootcampAnalyze modern code agents (Copilot, Cursor, Windsurf), compare transformer context windows vs RAG+AST systems.
lesson
Evaluating Preference AlignmentAI BootcampApply ensemble voting to eliminate hallucinations, compare DPO vs RLHF vs PPO pipelines (cost, control, complexity).
lesson
Preference-Based FinetuningAI BootcampUnderstand DPO, PPO, RLHF, GRPO; generate math-focused DPO datasets using numeric correctness.
lesson
Tool-Augmented FinetuningAI BootcampImplement two-stage finetuning (CoT → CoT+Tool), evaluate reasoning accuracy, and train quantized LoRA models.
lesson
Math Reasoning with SymPyAI BootcampUse SymPy for symbolic reasoning, fine-tune with Chain-of-Thought (CoT) data blending natural language and Python.