Curriculum Changelog

Track every update to the AI Engineering Bootcamp curriculum. We stay at the edge of AI engineering, constantly evolving with the latest tools, frameworks, and best practices.

Cohort 9 [v0.9]
current
Jan – Mar 2026
πŸ“œ View Detailed CurriculumDemo Day Projects: Coming Soon

All in on Agent Engineering.

New Additions

  • β€’πŸ§  Agent Memory
  • β€’πŸ“Ά Deep Agents
  • β€’πŸ•΅οΈΒ Dedicated Deep Research session
  • β€’1️⃣ Incorporated LangChain v1.0, LangGraph v1.0, and LangSmith Platform
  • β€’πŸš’ Clarity on types of servers (agent, LLM, app, MCP)
  • β€’πŸ”Œ Expanded MCP coverage on both client (connector) and server (agent communication) sides
  • β€’πŸŽ–οΈ Tougher Certification Challenge
  • β€’πŸ“Š Fully buzzword-compliant on Evals
  • β€’πŸ¦ Explicit Semantic Caching coverage

Deprecated

  • Γ—OpenAI Agents SDK
  • Γ—Dedicated On-Prem Session

Key Highlights

  • β†’Significantly more depth on agents, memory, deployment various levels of agent orchestration
  • β†’New primary cohort use case

Everything is context. RAG + Agents = Agentic RAG.

New Additions

  • β€’πŸͺŸ Expanded coverage and threading of context engineering
  • β€’πŸ’Ό How people are using ChatGPT and Claude
  • β€’πŸ—‚οΈ Classic RAG = Dense Vector Retrieval
  • β€’πŸŒ OSS models --> GPT-OSS and Embedding Gemma, remotely and locally
  • β€’πŸ“Š Adopting Evals terminology
  • β€’πŸ›€οΈ Combining Guardrails and Caching

Deprecated

  • Γ—Explicit coverage of LLM Ops

Key Highlights

  • β†’Demo Day Semi-Final Competition Round
  • β†’Primary cohort use case: What should you build and why?
  • β†’Early integration of open-source models in curriculum

Incorporating Context Engineering and agent communication

New Additions

  • β€’πŸͺŸ Context Engineering from first principles
  • β€’πŸ’Ό Primary cohort use case building throughout the cohort
  • β€’πŸ”Œ Expanded Model Context Protocol (MCP) content
  • β€’πŸ€– Agent2Agent (A2A) Protocol for building and running multi-agent systems
  • β€’πŸ›€οΈ Dedicated Guardrails session

Deprecated

  • Γ—Fine-Tuning of LLMs and Embedding models (now table stakes)
  • Γ—Inference & GPU Optimization

Key Highlights

  • β†’Primary cohort use case: Student loans customer service
  • β†’Protocol-first approach to agent architecture
  • β†’Enhanced security and guardrails implementation

Agent systems and production deployment foundations

New Additions

  • β€’πŸŽ² Increased prerequisites and an enhanced AIE challenge.
  • β€’πŸ§‘β€πŸ’» Cursor (with Claude), uv, and vibe coding! (Learn free)
  • β€’πŸ¦₯ Fine-tune Llama 3.1 for reasoning (Learn free)
  • β€’πŸ§ Build your own open-source Deep Research application with LangGraph (similar to this)
  • β€’πŸ”Ÿ Expanded 10-dimensional Learning Targets
  • β€’πŸ€© Certification Challenge even more aligned with Demo Day!
  • β€’πŸ›£οΈ Curated Journey Groups led by Peer Supporters.

Key Highlights

  • β†’First cohort with dedicated agent systems track
  • β†’Production deployment best practices established
  • β†’Comprehensive evaluation methodology introduced

Test-time compute and reasoning with enhanced peer support

New Additions

  • β€’πŸ€” Test-time Compute and Reasoning: It turns out that reasoning can be achieved within the LLM, or outside of the LLM through agentic patterns. In cohort 5, for the first time, we learn both methods of reasoning!
  • β€’πŸ§‘β€πŸ€β€πŸ§‘ Enhanced Peer Support: We've got an instructional staff of 10 experienced AI Makerspace-certified peer supporters here to elevate your learning experience. They'll lead breakout rooms, provide comprehensive grading feedback, and host a dedicated weekly office hour just for AIE5 students.
  • β€’πŸ₯½ Deep Dive Sessions: Back by popular demand, every Tuesday at 10 AM PT, join Vincent Kienzler for a 90-minute deep dive into the week's content, designed to clarify concepts and enhance your understanding.
  • β€’πŸ“… Midterm to Demo Day: We've redesigned the midterm to align with your Demo Day final project, ensuring a smoother, more impactful learning journey. Check out previous Demo Day projects!
  • β€’πŸŒŽ Alumni Meetups: Upon completing an AI Makerspace course, you'll become part of a global community of practitioners advancing their careers in AI Engineering. Join meetups, connect with peers, and continue growing with this thriving network.

Key Highlights

  • β†’First cohort to introduce dual reasoning approaches
  • β†’Expanded peer support with 10 certified instructors
  • β†’Weekly deep dive sessions for comprehensive learning

Dropped LlamaIndex to go all in on LangGraph

New Additions

  • β€’πŸ•΄οΈ Multi-agent applications with LangGraph.
  • β€’πŸ§ͺ Synthetic Data Generation (SDG).
  • β€’πŸͺ„ Advanced Prompt Engineering with DSPy.
  • β€’πŸ—‚οΈ Advanced RAG: semantic chunking, hybrid retrieval.
  • β€’πŸš€ Prod-readiness via caching, async, parallelization.
  • β€’πŸ’ On-premise considerations.
  • β€’βš›οΈ Inference optimization via quantization (GGUF & AWQ).
  • β€’πŸ”  More details on basic embedding models.
  • β€’πŸ’¬ Key front-end UI elements (e.g., PDF upload).

Deprecated

  • Γ—LlamaIndex (we've gone all in on πŸ”— LangChain).
  • Γ—OpenAI Assistants API (πŸ”— LangGraph).
  • Γ—WandB Prompts (πŸ”— LangSmith).
  • Γ—βš–οΈ Fine-Tuning: LlamaIndex HF Sentence Transformers.

Key Highlights

  • β†’βš–οΈ Fine-Tuning: HF Sentence Transformers.
  • β†’πŸš’ LLM serving & inference: vLLM.
  • →↗️ QDrant as go-to scalable vector DB.
  • β†’πŸ€– Chat model: GPT-4o-mini.
  • β†’πŸ€– Embedding model: text-3-embedding-small.
  • β†’πŸ¦Ύ Open chat model: Llama 3 8B Instruct.
  • β†’πŸ¦Ύ Open embeddings: Arctic Embed M.

Enhanced UIs, user feedback collection, and expanded project time

New Additions

  • β€’πŸ’¬ Building UIs for RAG & agentic systems
  • β€’πŸ—£οΈ Collecting and evaluating user feedback
  • β€’πŸ”’ Embedding fine-tuning before LLM fine-tuning
  • β€’πŸ—οΈ More time to build for Demo Day! (we've added a week!)

Key Highlights

  • β†’Focus on production-ready user interfaces
  • β†’Comprehensive feedback and evaluation systems
  • β†’Extended project development timeline

Comprehensive framework coverage with LangChain and LlamaIndex

New Additions

  • β€’πŸ€– No-Code GPT
  • β€’πŸ—οΈ LLM Application
  • β€’πŸ—ƒοΈ RAG Application
  • β€’πŸ•΄οΈ Agent Application
  • β€’βš–οΈ Fine-Tuned Model
  • β€’πŸ”— LangChain
  • β€’πŸ¦™ LlamaIndex
  • β€’πŸ“ˆ RAG Assessment

Key Highlights

  • β†’πŸ€‘ Industry Use Cases
  • β†’πŸš€ Open-Source Production RAG
  • β†’πŸ”„ Agentic RAG in Production
  • β†’πŸŒ Domain-Adapted RAG with Fine-Tuning

All in on AI Engineering!

New Additions

  • β€’The foundational bootcamp that started it all
  • β€’Comprehensive introduction to AI Engineering principles
  • β€’Building blocks for LLM applications
  • β€’Core RAG and agent concepts

Key Highlights

  • β†’Inaugural cohort establishing the curriculum foundation
  • β†’Introduction to production AI Engineering
  • β†’Community building and peer learning established

Want to join the next cohort and experience these updates firsthand?