AI Engineering - Build and Learn

Welcome to AI Build & Learn, a weekly AI engineering stream where we pick a new topic and learn by building together.

Up next: Open-Source Image Generation: Diffusion Models & LoRA (2026-07-10) Explore the best open-source image models, both diffusion and transformer-based, and try fine-tuning with LoRA. RSVP | Join the Slack

Upcoming:

🔨 Topic	📝 Description
Open-Source Image Generation: Diffusion Models & LoRA 2026-07-10 code • register	Explore the best open-source image models, both diffusion and transformer-based, and try fine-tuning with LoRA.
Open-Source Video Generation: Diffusion & Transformer Models 2026-07-17 code • register	Open-source text-to-video and image-to-video with diffusion and transformer/DiT models.
Open-Source Text-to-Speech: Natural Voices & Voice Cloning 2026-07-31 code • register	Turn text into speech with open-source TTS: natural voices, voice cloning, and multilingual output.
Open-Source Music Generation: Text-to-Music & Lyrics-to-Song 2026-08-14 code • register	Generate music and audio with open-source models: text-to-music and full lyrics-to-song with vocals.
Reinforcement Learning in MuJoCo 2026-08-21 code • register	Train reinforcement learning agents to control simulated robots in MuJoCo, the open-source physics engine used as a standard RL benchmark.
Reinforcement Learning in NVIDIA Isaac Sim: Physical AI & Sim-to-Real TBD code	Step up from MuJoCo to NVIDIA Isaac Sim: GPU-accelerated, photorealistic robot simulation for physical AI and sim-to-real transfer.
World Models with DreamerV3 TBD code	An intro to world models: DreamerV3 learns a compact model of its environment and trains a policy inside imagined rollouts. Lightweight and RL-connected.
World Models with NVIDIA Cosmos: Physical AI TBD code	NVIDIA Cosmos, an open world foundation model for physical AI: generative world simulation and action-conditioned rollouts, runnable locally on a DGX Spark.
World Models with V-JEPA 2: Latent-Space Prediction & Planning TBD code	A different take on world models: V-JEPA 2 predicts in representation space instead of pixels. Self-supervised video understanding and latent planning, runs on a DGX Spark.

Past:

🔨 Topic	📝 Description
MLflow: ML Tracking, LLM Tracing & LLM-as-a-Judge 2026-06-26 code • watch	MLflow for the full ML lifecycle: experiment tracking, model registry & evaluation, LLM tracing, prompt registry, and LLM-as-a-judge, all running on Flyte.
Arize Phoenix: LLM Observability 2026-06-12 code • watch	Arize Phoenix is an open-source observability tool for LLM apps: trace, evaluate, and debug your RAG and agent pipelines.
Ragas: Evals for RAG & Memory 2026-06-05 code • watch	Ragas evaluates RAG and memory systems with structured metrics: retrieval quality, faithfulness, answer relevance, and more.
Cognee: Memory Layer for Agents 2026-05-22 code • watch	Cognee is an open-source memory layer for AI agents. Persistent, queryable memory that survives across sessions and tools.
Karpathy's LLM Wiki 2026-05-15 code • watch	Andrej Karpathy's LLM Wiki concept: build a wiki-style knowledge base specifically structured for LLM consumption and retrieval.
Graph Data with Neo4j 2026-05-08 code • watch	Graph databases with Neo4j for GraphRAG and agent context. When to reach for knowledge graphs, and how to combine graph traversal with semantic search.
Vector Stores 2026-05-01 code • watch	Embeddings and vector stores: how semantic search actually works under the hood and how to build RAG pipelines on top.
Gemma 4 2026-04-24 code • watch	Build with Google's newly released Gemma 4 models across different sizes for chat, agents, and visual understanding.
AutoResearch 2026-04-17 code • watch	Andrej Karpathy's AutoResearch concept: give an AI agent a small but real LLM training setup and let it experiment autonomously overnight.
Reinforcement Learning with OpenEnv 2026-04-10 code • watch	An e2e framework for creating, deploying and using isolated execution environments for agentic RL training, built using Gymnasium style simple APIs.
Agentic Search with Tavily 2026-04-03 code • watch	Tavily is a search API often used by AI agents. Learn how to integrate Tavily search into your AI applications.
MCP with FastMCP 2026-03-27 code • watch	Learn what MCP (Model Context Protocol) is and how to build an MCP server using FastMCP.

Setup

# Clone the repository

# Create virtual environment
uv venv .venv --python 3.11

# Activate the venv
source .venv/bin/activate  # macOS/Linux
# or
.venv\Scripts\activate     # Windows

# Install dependencies
uv pip install -r TopicFOLDER/requirements.txt