Skip to main content

​
New to AIDO NOW?

If you’re exploring how AI can transform your development workflow, you’re in the right place. This site documents real learnings from building production systems with multi-agent AI - the successes, the failures, and the practical patterns that emerged. Choose your path based on what you need right now.

​
Three Learning Paths

Path 1: New to AI Development

Start with fundamentals and build understanding progressively

Path 2: Need Immediate Results

Get practical tools you can use today

Path 3: Full Journey

Follow the complete chronological experience

​
Path 1: New to AI Development

Best for: Engineers exploring AI workflows for the first timeTime commitment: 45-60 minutesYou’ll learn: How to structure AI collaboration effectively
1

Start: Multi-Agent Workflow

Understand why single-prompt coding fails and how Evaluator-Builder-Verifier agents transform development.

Multi-Agent AI Workflow

Why you need separate agents for planning, building, and verification
Key takeaway: Fresh sessions for verification catch 70% more bugs than reusing the same session.
2

Then: When AI Excels

See where AI provides genuine value beyond the hype.

When AI Excels

How AI transforms tedious work into valuable output
Key takeaway: AI makes previously “not worth it” work suddenly feasible - like comprehensive documentation and systematic testing.
3

Finally: When AI Fails

Learn the boundaries and failure modes before you encounter them.

When AI Fails: Cascading Errors

30 commits to fix one change - what went wrong and how to avoid it
Key takeaway: Macro changes and breaking changes don’t work well with AI’s fix-in-session approach.

​
Path 2: Need Immediate Results

Best for: Developers who want practical tools nowTime commitment: 30 minutesYou’ll get: Copy-paste prompts and decision frameworks
1

Start: Prompt Library

Get tested prompts for infrastructure, debugging, and code review.

The Prompt Library I Actually Use

Copy-paste ready prompts tested over 7 weeks of production work
Immediate value: Infrastructure design, database schema design, code review prompts with 90%+ success rates.
2

Then: AI Tool Comparison

Choose the right tool for your specific task.

Claude vs GPT-4 vs Copilot

Opinionated guide based on real usage, not benchmarks
Immediate value: Decision tree for when to use which tool - save time and money.
3

Finally: AI Limitations

Avoid wasting time on tasks AI can’t solve.

AI Limitations Boundary

What AI genuinely can’t do (and what humans must own)
Immediate value: Clear boundaries prevent frustration and wasted effort.

​
Path 3: Full Journey

Best for: Those who want the complete story with contextTime commitment: 3-4 hoursYou’ll experience: The full evolution from week 1 to week 7
1

Start: Journey Overview

Understand the experiment and what you’ll learn.

Building with AI - Series Overview

The 7-week journey using multi-agent AI to build production systems
What you’ll get: Context for the entire series, key themes, metrics that matter.
2

Follow Chronologically

Read weeks 1-7 in order to see patterns emerge.

Week 1: Multi-Agent Setup

Setting up Evaluator, Builder, Verifier workflow

Week 2: Plan-Implement-Verify

Three-phase workflow with quality gates

Week 3: AI Event Sourcing

How AI helped (and hindered) event sourcing design

Week 4: When AI Excels

Boilerplate generation and systematic work wins

Week 5: When AI Fails

Architecture mistakes and cascading errors

Week 6: AWS Runtime Adoption

Breaking changes and macro evolution

Week 7: Config Governance

Middleware patterns and systematic implementation
3

Deep Dive by Topic

After chronological reading, explore specific interests.AI Practices:When AI Works:When AI Struggles:Practical Tools:

​
Not Sure Which Path?

Question 1: How much time do you have right now?
  • Less than 30 minutes → Path 2 (Immediate Results)
  • 45-60 minutes → Path 1 (New to AI Development)
  • Multiple hours → Path 3 (Full Journey)
Question 2: What’s your primary goal?
  • Learn AI workflows → Path 1
  • Get tools I can use today → Path 2
  • Understand the full context → Path 3
Question 3: How do you prefer to learn?
  • Progressive building blocks → Path 1
  • Practical, hands-on tools → Path 2
  • Complete narrative arc → Path 3
Question 4: What’s your AI experience level?
  • Never used AI for coding → Path 1
  • Used Copilot, want to do more → Path 2
  • Serious about AI workflows → Path 3
Based on what resonates with readers exploring AI development workflows.

Multi-Agent AI Workflow

Why single-prompt coding doesn’t scale and how three specialized agents transform development

The Prompt Library

Copy-paste ready prompts for infrastructure, debugging, and code review - tested over 7 weeks

When AI Excels

How AI transforms tedious systematic work into valuable output that would never get done manually

When AI Fails: Cascading Errors

30 commits to fix one change - understanding AI’s limitations with breaking changes

Claude vs GPT-4 vs Copilot

What I use when - opinionated guide based on real usage for infrastructure and development

​
After You Read

Subscribe for Updates

Get new articles delivered weekly as I continue exploring AI workflows

About This Work

Learn about the approach and philosophy behind this content
All content reflects real learnings from personal projects. Your results may vary based on your context, tools, and constraints.