Build Lab | May 21, 2026

Intelligence Per Dollar

How to get more useful work from every AI dollar

Download as Markdown View Presentation

The Core Formula

Intelligence Per Dollar = (Value × Throughput) ÷ Cost

Most people obsess over cost alone. The real game is increasing value and throughput while keeping cost reasonable.

Three Levers:

Increase value — Get more useful work per token
Increase throughput — Process more work efficiently
Decrease wasted cost — Stop paying for tokens that don't deliver

5 Myths, Busted

The things people get wrong about AI.

MYTH 1

"Frontier vs. open source is the decision"

→ Use both. Frontier for planning, open source for execution.

MYTH 2

"Buy a Mac Mini, get frontier smarts"

→ Local is great for learning. Frontier is still significantly smarter for serious work.

MYTH 3

"Big context = big cost"

→ Caching changes the math. Cached input is 40× cheaper than output.

MYTH 4

"Ask AI which open model to use"

→ Check live benchmarks. Open source moves faster than training data.

MYTH 5

"Need expensive models for great results"

→ Systems beat model size. Architecture > raw power.

Agent Architecture Patterns

The patterns that actually work.

HAHA

Human-Agent-Human-Agent

H → A → H → A

Humans are the quality layer at both ends. Best for: Work requiring human judgment + AI iteration.

HAaAH

Human-Big-small-small-Big-Human

H → A → a → a → A → H

Big models handle reasoning-heavy parts (start/end). Small models do bulk work (middle). Best for: Cost-effective high-quality output.

THE SYSTEM LOOP

Generate → Verify → Revise → Retry

Shows up everywhere: Google Aletheia, Poetiq, Microsoft MDASH (88.4% on CyberGym vs Claude Mythos 83.1%).

The insight

You don't need bigger models. You need better systems.

The Both Strategy

How to use frontier + open source together.

Frontier for thinking (planning, architecture, complex reasoning).
Open Source for doing (execution, implementation, high-throughput work).

STEP 1: Plan with Frontier

"Plan this feature refactoring. Write your plan as if passing to a junior engineer. Include all context they need to execute."

STEP 2: Execute with Open Source

"Implement this plan. Write clean code. Add tests for all edge cases."

Result: Strong output at a fraction of the cost.

The Double Hack

Four models. One task.

This is the generator-verifier pattern from MDASH/Aletheia, applied to your actual work.

FRONTIER (Claude/GPT)

"Write a spec/plan as if handing off to a junior engineer"

OPEN SOURCE (GLM)

"Implement this entire plan"

OPEN SOURCE (GLM)

"Verify the implementation. Run tests."

FRONTIER (Claude/GPT)

"Code review. Any issues?"

The /goal Pattern

The simplest agentic pattern.

Define completion, let AI iterate until done.

/goal Refactor auth into separate module, ensure all tests pass, add 3 new tests

How it works:

You define completion condition
Agent works on task
Evaluator checks: Are we done?
Loop until YES

Great for: Refactoring, test coverage, debugging, documentation, competitor research, customer feedback analysis.

Tools & Services

What to use and where to get it.

GLM (Zhipu AI)

TOP PICK FOR CODING

Lite

$14/mo

Small repos

Pro

$58/mo

Day-to-day

Max

$128/mo

Heavy use

Kimi (Moonshot AI)

RESEARCH & LONG CONTEXT

Moderato

$15/mo

Basic

Allegretto

$31/mo

Pro users

Allegro

$79/mo

Premium

ArtificialAnalysis.ai — Live model comparisons. Always check before choosing.

RESEARCH & READING

Google Aletheia Paper: arxiv.org/html/2602.10177v3 — Generator → Verifier → Reviser pattern

Microsoft MDASH: Official Security Blog

Poetiq: Recursive Self-Improvement Explained

Multi-Agent Frameworks: Archon (orchestration), Paperclip (prompt version control), Agent Zero (autonomous agents)

Prompts to Give Your AI

Copy these. Use them tonight.

Set Up a HAHA Workflow

"Set up a workflow where:
1. I provide initial requirements and verify the final output
2. You generate a first pass
3. I review and provide feedback
4. You revise based on my feedback
5. We iterate until I approve

Work on [TASK] using this pattern."

Set Up a HAaAH Double-Hack

"I want to use a 4-step model workflow for cost efficiency:
1. Use a frontier model (Claude/GPT) to plan this task: [TASK]
2. Use an open source model (GLM) to implement the plan
3. Use the same open source model to verify the implementation
4. Use a frontier model to do a final code review

Set up this workflow for me."

Create a /goal-Style Agent

"Create an agent that:
1. Takes a goal as input: [GOAL]
2. Works toward the goal autonomously
3. Self-checks if the goal is complete
4. Iterates until completion criteria are met

The goal is: [YOUR GOAL]"

Tonight's Checklist

BEFORE YOU SLEEP

Try /goal on ONE real task
Check ArtificialAnalysis.ai for current model rankings
Share one experiment with someone from the talk
Pick ONE workflow to systematize

Small experiment. Real task. Tonight.

Key Takeaways

Systems > Scale — Architecture beats model size
Use both — Frontier for thinking, open source for doing
Cache aggressively — Cached input is 40× cheaper than output
Human-in-the-loop — U + AI beats AI alone
Start simple — /goal is your entry point

You don't need bigger models.
You need better systems.

For Business Owners

Design your own multi-agent workflows.

Don't just copy workflows. Learn to design them.

START HERE

If you run a team, pick one workflow that is:

Repetitive — It happens every week
Slow or expensive — It costs real time, money, or attention
Important enough to matter — Better output would change something
Easy for a human to verify — A manager, expert, or owner can tell if it worked

Then paste the Workflow Designer Prompt below.

For your first /goal, make the task measurable, scoped, and self-served.

THE DESIGN FRAMEWORK

Identify — What repetitive work needs automation?
Decompose — Break it into discrete steps
Assign — Frontier (thinking) or Open Source (doing)?
Connect — How does output flow between steps?
Verify — Where do humans check quality?

Patterns: HAHA (Human→AI→Human→AI) for judgment-heavy work, or HAaAH (Human→Big→small→small→Big→Human) for cost-effective quality.

THE WORKFLOW DESIGNER PROMPT

Copy this prompt and paste it into Claude or GPT. The AI will guide you through designing your own workflow, one question at a time.

I want to design my first multi-agent workflow for my team.

Act as a workflow design consultant. Your job is to help me pick a good first workflow, design it, and turn it into a small pilot we can run this week.

Ask questions ONE AT A TIME. Wait for my answer before asking the next question. If my answer is vague, give me 2-3 examples and ask me to choose.

PHASE 1: Understand the Current State
Start by learning about my situation:
1. What repetitive work does my team do every week?
2. Who currently does this work and how long does it take?
3. Where do mistakes or bottlenecks typically happen?
4. What would success look like for this workflow?
5. How will a human verify the output is good?
6. What assets, documents, templates, or examples already exist?
7. What tools or systems can AI access, and what must stay manual?
8. What channels or platforms are involved?
9. Which parts require a specific human to decide, approve, record, or build trust?

PHASE 2: Decompose the Work
Once you understand my situation, help me:
- Break the work into discrete, sequential steps
- Identify which steps require creative thinking vs. execution
- Flag any steps that require human judgment (domain expertise, compliance, etc.)
- Note any steps that could run in parallel
- Separate AI technical sprints from human real-world handshakes
- If the outcome is bigger than one sprint, break it into a 3-4 week mission with mini-goals

PHASE 3: Assign Models to Steps
For each step, recommend:
- Frontier model (Claude/GPT) for: planning, architecture, complex reasoning, final review
- Open source model (GLM/Kimi) for: execution, implementation, data processing, verification
- Human verification for: quality gates, sensitive decisions, brand voice, compliance
- Manual handoff if we do not have sub-agent tooling yet

PHASE 4: Choose the Pattern
Based on the work, suggest which pattern fits best:
- HAHA (Human→AI→Human→AI): Humans at both ends, best for judgment-heavy work
- HAaAH (Human→Big→small→small→Big→Human): Frontier for planning/review, open source for execution
- Simple loop: Single AI with human verification
- Manual multi-model workflow: Copy outputs between models when tools do not support sub-agents

PHASE 5: Design the Flow
Map out how information flows between steps:
- What output does each step produce?
- What does the next step need as input?
- Where should we cache or store intermediate results?
- What's the verification/feedback loop?

PHASE 6: Implementation Plan
Finally, give me:
- A step-by-step workflow I can implement
- Specific prompts for each step
- How to test this with a small pilot
- Metrics to track to know if it's working
- Risks and failure modes to watch for
- The simplest no-code/manual version of the workflow
- The more automated version if my tools support sub-agents or agent teams

Final output format:
- Workflow name
- Business goal
- Goal type: single sprint or multi-week mission
- Steps
- Model or person assigned to each step
- Human verification points
- Human-only action items
- Copy-paste prompts
- First pilot plan
- Success metrics
- Known risks

Ask me the first question now.

This prompt teaches you WHILE it helps you.

EXAMPLE: HOW IT WORKS

AI asks:

"What repetitive work do you want to automate?"

You say:

"Support agent reads email, checks policy, verifies purchase, drafts response"

AI asks:

"Where do mistakes happen?"

You say:

"Perfect, let's try it"

COMMON WORKFLOWS TO CONSIDER

Workflow	Pattern	Why
Customer support triage	HAHA	AI drafts, human approves for quality
Report generation	HAaAH	Frontier plans, GLM crunches data, human reviews
Code review	HAaAH	GLM scans, frontier flags issues, human decides
Meeting summaries	HAHA	AI drafts summary, human checks accuracy
Documentation updates	HAHA	AI drafts, human verifies technical correctness

FIRST STEPS

Pick ONE repetitive workflow your team does
Use the Workflow Designer Prompt with your AI
Run it as a pilot with 2-3 people
Measure the difference — time saved, quality maintained
Expand what works

This scales beyond your team

Team training, workflow architecture, cost optimization audits, AI system design.

Talk to us after if you're running a business and want to scale these patterns.

← Back to Allenix

#IntelligencePerDollar • Build Lab | May 21, 2026