Geode Labs · Build in public
The open build and research behind Yu.
Geode Labs is building Yu: a private place a person fills with the pieces of their own thinking, and an AI assistant that helps them use what they have kept. This page is the working record of that build. It changes as the evidence changes, and it says plainly what is still unproven.
The durable system is a 52-week method: one piece of context at a time, reviewed and approved by the person before anything is kept. The 21-step build running today is the pilot onboarding wrapper around the first weeks of that method, not the product itself.
While I was building Yu with AI, I started using AI to catalog the build itself. This page is the public view of that record: what changed, what the evidence says, what I am testing, and what comes next.
What is live now
Yu runs end to end.
A week ago, these were pieces of Yu. Now they are one system.
Latest meaningful change
Onboarding, profile/reflection, the 21-step first build, progression, AI 101, the city, weekly delivery, AI handoff, and assistant arrival now run as one end-to-end system rather than separate product pieces.
452+
recorded build edits since Aug 31
Implementation activity, not company or strategy changes. Captured during the active sprint and refreshed when it closes.
150
in the final ~28 hours
21
first-build activities now connected into one progression path
Next week
beta testing starts
Beta testing starts next week, not today.
Build edits are implementation activity. They are separate from the documented company/product/strategy state changes counted below, and the two should not be read as the same thing.
Current sprint
V4 integration sprint
Remove the upstream comprehension, motivation, progression, and integration blockers so the methodology can actually be tested.
Current build focus
V4 integration sprint: connecting onboarding, profile/reflection, the 21-step first build, progression, AI 101, the city, weekly delivery, AI handoff, and assistant arrival into one end-to-end system so the methodology can actually be tested.
Current result
Onboarding, profile/reflection, the 21-step first build, progression, AI 101, the city, weekly delivery, AI handoff, and assistant arrival now operate as one end-to-end system rather than separate product pieces.
Active experiment
The question we are trying to answer next.
Open question
Can real beta users understand the value, start, stay engaged, complete enough of the first 21-step path, and reach the methodology with low correction and confusion?
Next test
Run the first beta test through the full 21-step path, then compare generic AI vs self-report/custom instructions vs Yu-derived translated instructions on genuinely held-out tasks.
7 dated milestones · Apr 21, 2026 to Sept 9, 2026
How the build moved.
Yu has developed by repeatedly finding the uncertainty that prevents the next important question from being answered.
Working readiness values below are a dated internal assessment, not an external certification. No outside body awarded them.
- V1
Manual Yu
Working TRL 2 → 3Question
Can we actually do cognitive translation?
What existed
Paisley + Danny were effectively the system. They used a framework generator, ran the conversations, interpreted what came back, built their own system instructions, and created individualized scaffolding for other people.
What this established: Enough manual proof of mechanism to justify productizing it.
- V2
Make the idea legible
Working TRL ~3TRL barely moved, and this stage still mattered.
Question
Can people understand what Yu is and why it matters?
What existed
The marketing site, positioning work, explanation of the six applied areas, and the language and visual system for explaining Yu.
What this established: Explaining the mechanism and value was itself a product problem.
- V3
First framework MVP
Working TRL 4 → 5Intended question
Does the methodology work when users go through Yu themselves?
What happened instead
People were not reliably getting far enough through the experience to answer that question properly.
What blocked the question
Perceived chore
Response: Gamification, progression, city, rewards, visible accumulation.
Value / concepts not landing
Response: AI 101, clearer conceptual scaffolding, stronger explanation of why the work matters.
Key interpretation: The methodology test was blocked by an upstream comprehension and motivation problem. That blocker became the new critical path.
- V4
Integration sprint
Working TRL 5 → 6Working estimate.
Sprint objective
Can we remove the upstream comprehension, motivation, progression, and integration blockers so the methodology can actually be tested?
Experience track
Understand the value → want to start → stay engaged → reach the frameworks.
Method track
Strengthen the underlying methodology through research and expert and psychologist review, so what people reach is defensible.
What this established: Onboarding, profile/reflection, the 21-step first build, progression, AI 101, the city, weekly delivery, AI handoff, and assistant arrival now operate as one end-to-end system rather than separate product pieces. Yu will be ready for beta testing next week.
Next readiness question
Does Yu's translated understanding measurably improve real AI use?
Dated chronology
Apr 21, 2026
The first documented precursor question
Should knowing how I think change how AI works with me?
May 11, 2026
First technical architecture
Jun 3, 2026
Mirror identified
A missing functional step between inference and instruction.
Jul 3, 2026
Raw-source and provenance architecture formalized
Aug 2026
Framework library expanded and sequenced
The methodology grew into an ordered library.
Sept 2026
Working product and research scaffold
Conditional instructions and misfire architecture, the cognitive city, and a formal research scaffold.
Sept 9, 2026
First 21-step path connected end to end
Onboarding through assistant arrival now run as one system. Yu will be ready for beta testing next week.
102
documented company/product/strategy changes
Business and product evolution. Compiled through Sept 9, 2026.
4
true reversals
Compiled through Sept 9, 2026.
The meaning matters more than the number: most changes narrowed and refined the model rather than replacing it. These are company/product/strategy state changes, separate from the build-edit count in the current sprint above.
Assessed Sept 9, 2026
Working readiness assessment
A dated internal working assessment from the Yu journey-map evidence review. These are not external certifications and no outside body awarded them.
Technology Readiness Level
Working assessmentTRL 6
Technology demonstrated in a relevant environment.
1
2
3
4
5
6
7
8
9
The assessment placed Yu at TRL 6 because the full pipeline runs end to end against real external users on a production-shaped stack. It deliberately did not award TRL 7, because the system was not yet at or near operational scale in the target environment.
Caution: The missing held-out accuracy test is the biggest technical uncertainty that could change this assessment.
Innovation Readiness Level
Working assessmentIRL 4
Prototype a low-fidelity minimum viable product.
1
2
3
4
5
6
7
8
9
The assessment placed Yu at IRL 4 because user activation and willingness-to-pay evidence were still early while Yu is free to use. That is the current evidence boundary and the next thing the company is designed to learn.
2-level readiness gap
Product build has advanced faster than paid-market evidence. The next experiments are designed to close the evidence gap.
What moves the levels?
TRL: Held-out accuracy and predictive-validity evidence, plus an operational-scale environment.
IRL: Paid use, activation, retention, and willingness-to-pay evidence.
These are evidence thresholds, not promises.
Readiness of the evidence
What we can stand behind, and what we cannot.
Supported
3Backed by the current architecture, the shipped build, or established outside research.
Person-specific context can matter to AI personalization.
Yu has a working behavior-informed assessment and translation pipeline.
The product preserves source and provenance, and the person can correct what it holds.
Testing
3An open hypothesis with a test designed for it.
Behaviorally derived instructions reduce correction burden compared with generic AI or self-written custom instructions.
Translated directives transfer better across tasks.
Conditional instructions reduce misfires.
Unknown
3An important question we cannot answer yet.
Which framework categories add the most incremental value.
How stable different findings are over time.
How well the benefits transfer across model providers.
SUPPORTED means supported by the current architecture, the shipped implementation, or established external research. It does not mean Yu has been scientifically validated. The framework categories are Geode Labs' working applied taxonomy for organising the method, not a validated model of the mind.
Build log
What changed, and when.
Sept 9, 2026
V4 integration sprint
Onboarding, profile/reflection, the 21-step first build, progression, AI 101, the city, weekly delivery, AI handoff, and assistant arrival now run as one end-to-end system rather than separate product pieces. Yu will be ready for beta testing next week.
Sept 2026
City simplification
The cognitive city was simplified so every place has one clear job and the map reads at a glance.
Sept 2026
Conditional instruction and misfire design
Instructions gained conditions: they know when they apply and stay quiet when they do not.
Sept 2026
Public methodology and taxonomy research scaffold
The working paper and this build record opened the method up to outside scrutiny.
Methodology
The working paper.
The method is written up so it can be read and argued with, not just described in marketing copy.
Cognitive Translation for Personalized AI
A Taxonomy and Methodology for Modeling Person-Specific Context
Working paper · under active validation
Part of the work