Foundations 2.0

From AI User to
Agentic Architect

Six weeks. Live sessions where Tyler builds on camera, mistakes and all. By the end, you have a working multi-agent system you built yourself and can prove it works. No code required.

4.6/5
400+ Reviews
1,700+
Graduates
17
Cohorts Shipped
15+
Countries
Join the Next Cohort →
Who This Is For
You don't need to be technical.
If any of these sound like you, this course was built for you.
💼

The Operator

You run a business or a team. You know AI should be doing more of your work, but you haven't figured out how to make it actually stick.

💡

The Founder

You've used ChatGPT but haven't built anything real. You want to go from consumer to builder and stop paying others to do what you could own.

🚀

The Career Mover

You want "AI" on your resume for real. Not a certificate from a video course. An actual production system you built and can prove works.

What You Build
Not tutorials. A production system.
By graduation, you've built, tested, and shipped a complete multi-agent system that handles real work. Here's what's running when you're done.
🤖

6 Specialized Agents

Expert, research, sentiment, QA, routing, and custom specialist agents, each with their own knowledge base and instructions.

⚡

Dual Workflows

Email Handler (autonomous processing) and Human Feedback Loop (oversight). Intelligent routing between agents with safety nets.

📈

Evaluation Evidence

Structured evaluation showing 80%+ performance, before-and-after improvement metrics, and ROI calculation proving business value.

The Curriculum
6 weeks. 36 modules. All hands-on.
Every week builds on the last. You learn a concept, then immediately build with it. Nothing theoretical stays theoretical.
  • The Agentic Shift // Chatbot vs Agent vs Agentic System. The paradigm shift from "chat" to "workflows."
  • Mechanics of Intelligence // How LLMs work (practitioner-level). Tokens, context windows, attention, temperature, output formats.
  • Context Engineering // The difference between prompt engineering (what you say) and context engineering (what the model knows). Seven components of agent context architecture.
  • Deconstructing The Professor // How system instructions shape AI behavior. Analyzing a production agent live.
  • Intro to RAG // Retrieval-Augmented Generation: giving agents access to custom knowledge bases.
What you build

Your own personalized "Professor" AI assistant with custom system instructions tuned to your voice, role, and use case.

  • The Autonomy-Reliability Connection // Autonomy must be earned through demonstrated reliability. Three failure modes, four factors of trust.
  • Designing Your Agentic Department // Workflow-first thinking: design the system before building agents.
  • The Fisk Method // 8-phase development process for building production-ready agents with AI as your co-builder.
  • AI Agent Blueprint // Knowledge, Tools, Inputs, Outputs, Constraints. The framework for every agent you'll ever build.
  • Echo Voice Analysis // Automated analysis of your communication style so your agents write like you.
  • TOAST Testing // Take, Observe, Analyze, Stick, Test. Systematic agent testing methodology.
What you build

"You've Got Mail": a production email agent connected to a knowledge base, configured with your voice, tested against rubrics, and demonstrating 80%+ reliability.

  • Utility Agents // Cinnamon (sentiment analysis) and Kansa (quality assurance). The support layer every agent team needs.
  • Building Hatch // Your domain expert agent. Deep knowledge, RAG-connected, evaluated to 80%+ performance.
  • Building Scratch // Your web research agent. External information gathering integrated into your workflow.
  • Building Holler // Your router/orchestration agent. Intelligent routing and decision logic for multi-path workflows.
  • RCICE Framework // Role, Context, Instructions, Criteria, Examples: the structure for every system instruction you write.
What you build

Complete 6-agent team: Hatch (expert), Scratch (research), Holler (router), a custom specialist, plus Cinnamon (sentiment) and Kansa (QA).

  • Human-In-The-Loop // Why humans are part of production systems. Circuit breaker pattern and safety nets before external actions.
  • User Prompts for Assembly // The difference between system instructions (job descriptions) and user prompts (work orders). 40+ prompts across triage, research, and response.
  • Workflow Wiring // Step-by-step assembly of your Email Handler and Human Feedback Loop workflows.
  • Production Testing // End-to-end testing of the complete system with real inputs.
What you build

Two production workflows: Email Handler (autonomous processing from intake to delivery) and Human Feedback Loop (escalation, review, correction). Your system handles real work.

  • Core Eval Framework // Two-level evaluation: step-level (individual agents) and run-level (complete system). Traffic light scoring: Pass, Review, Fail.
  • Meet Evy // Your AI evaluation partner for calibrating your scoring against consistent standards.
  • Targeted Improvement // TOAST methodology applied: identify the single most impactful fix, implement, and prove improvement with data.
  • Calculating ROI // Real business impact: time saved, quality improved, cost reduced. Metrics specific to your use case.
What you build

Comprehensive evaluation report showing 80%+ performance, before-and-after improvement metrics, ROI calculation, and your official AI Build Lab Agentic Practitioner Certificate.

  • Agent Build Week // Dedicated time to build, iterate, and polish your complete system. Office hours and peer support throughout.
  • Show and Tell // Present your system to the cohort. See what your classmates built. Get live feedback from Tyler and Sara.
  • Graduation + What's Next // Receive your certificate. Connect with the alumni community. Preview Summer Camp and ongoing opportunities.
What you build

Your polished, production-ready capstone: a complete multi-agent system with evaluation evidence and ROI metrics. Plus the connections, confidence, and certificate to take it forward.

Sound like what you've been looking for?
Join the Next Cohort →
Proprietary Frameworks
Built from the field, not the classroom.
These are the frameworks Tyler and Sara use in their own client work. You learn them in week one, then use them every week after.
The Fisk Method

8-phase development process for building production-ready agents using AI as your co-builder. The core methodology behind every agent you build in the course.

RCICE Framework

Role, Context, Instructions, Criteria, Examples. The structure for writing system instructions that produce reliable, consistent agent behavior.

TOAST Method

Take, Observe, Analyze, Stick, Test. Systematic agent testing and improvement methodology. Identify the single most impactful fix and prove it with data.

AI Agent Blueprint

Knowledge, Tools, Inputs, Outputs, Constraints. Five components that define every production AI agent. The template you'll use for every agent you ever build.

36
Modules
6
Agents Built
2
Workflows
6
Weeks
Results
What our graduates are doing.
These are people who showed up for the same 6 weeks you're considering.
"OFFICIALLY AN AI CONSULTANT!! Tomorrow I wrap up my first week as the fractional CAIO for a national public health organization at Georgetown University."
Paige Bradbury
Fractional CAIO, Georgetown University
"We just closed our first client yesterday, officially signing the SOW for TIP. Even better, on the same day, we submitted our first 6-figure proposal."
Kim Siler
Co-Founder, TIP
"Hands down the best course I've ever taken. Sara and Tyler are extremely knowledgeable, open-hearted, and fun."
Dominique Neff
Director, Product Management // Starz
"Today Eclipse AI is showcasing its AI Agent Swarm at the ServiceNow Public Sector Forum! Advanced discussions with three massive customers, a $3.8bn manufacturing business..."
Chris CJ Jones
Founder, Eclipse AI
After Graduation
The course ends. The community doesn't.
Foundations is 6 weeks. But what you get access to lasts a lot longer.
👥

Alumni Community

1,700+ alumni in a private Slack. Active channels, peer support, job leads, and collaboration. No lurkers, just builders.

⛰

Summer Camp

An ongoing experience for grads. New blueprints, live build sessions, show and tell, guest experts. Launching summer 2026.

🎓

Certificate + Portfolio

Official AI Build Lab Agentic Practitioner Certificate with a verifiable credential. Plus a working system you can show clients or employers.

Questions
Common questions.
No. Foundations 2.0 is designed for non-technical operators. You build real AI agent systems using no-code tools. If you can write an email, you can take this course.
A computer with internet access. The primary platform is Cassidy.ai (included). You'll also use Claude, ChatGPT, and other tools we provide access to during the course.
Plan for 6-8 hours per week. That includes live sessions (2-3 hours), async video modules (1-2 hours), and hands-on building time (2-3 hours). Some weeks are heavier during build phases.
All live sessions are recorded and available within 24 hours. You can catch up async. The Slack community is also active, so you can get help between sessions.
No. Prompt engineering is one small piece of what we teach (we call it context engineering). Foundations 2.0 goes way beyond prompting to teach system architecture, multi-agent orchestration, workflow assembly, and structured evaluation. You don't just learn to talk to AI. You learn to build teams of AI agents that work together autonomously.
A working multi-agent system you can use at work or sell to clients. Evaluation evidence proving it works. An ROI calculation showing business value. An official AI Build Lab Agentic Practitioner Certificate with a verifiable credential. And access to the alumni community, ongoing events, and Summer Camp.

Ready to build?

Your first AI agent ships in Week 1. By Week 6, you're running a production multi-agent system. No code required. Bring 6-8 hours a week and a use case you want to automate.

Join the Next Cohort →
LIVE COHORT // 6 WEEKS // NO CODE REQUIRED