The Infinity Method: A Practitioner’s Grammar for Finding Focus in Friction
This isn’t a manual for AI tools. It’s a grammar for seeing your own work more clearly. It’s a way to find the signal in the noise of your daily operations and to use technology not as a mandate, but as a partner in creating more meaningful, valuable work.
Most conversations about AI start with the technology—the “vendor narrative.” They want to sell you a solution before you’ve even named your problem. We don’t do that. We start somewhere much more real.
“You really have to start from the inside… you really have to understand what am I even looking for? Like what, what, how do I even find a problem that’s a good fit for this?”
We start with the work itself. With the “sand in the gears” of your business. We start with friction.
1. See the Friction, Then Define the Entire Opportunity
Friction isn’t a failure. It’s a compass. It points directly to the places where your team is frustrated, where work gets stuck, or where customers feel a delay. It’s the most honest signal you have for where to find your next strategic advantage.
Once you see the friction, you must define the value you intend to create. We use the SEA framework not to pick one lane, but to set a higher bar for what a true solution looks like.
- Streamline: Does this make the work easier for our people?
- Enrich: Does this make the output better for our customers?
- Accelerate: Does this make the process faster for everyone?
The challenge is to find a solution that does all three. That’s where the real transformation lies.
“If you haven’t yet figured out how to get all three from what you’re doing, you didn’t fix, you didn’t finish looking at the opportunity, right? So all you did was sped it up, or all you did was got a better result, or all you did was reduce the effort, you didn’t quite crack the nut yet, right?”
2. Map the Value Chain: Make Your Intuition Explicit
Before you can improve a process, you have to honor it. As founders and experts, you have an intuitive understanding of how your business creates value. The goal here is to make that intuition explicit.
“When you see a problem that you know how to solve, you’re like, okay, this is how I solve this problem. These are the steps I’m going to take. Even if you don’t literally have that written down somewhere, you just intuitively know… that is a knowledge workflow.”
We map this “knowledge workflow” by telling its story. A great way to start is by working from both ends toward the middle:
- Start with the customer: What is their lived experience when you deliver perfectly? What pain went away?
- Then, start with your advantage: What is the unique thing you or your company does at the beginning of the chain that makes this value possible?
- Fill in the middle: What are the core steps that connect your advantage to their experience?
This map gives us a shared definition of success and a stable foundation to stand on before we introduce any change.
3. Scope the Work: What’s Possible vs. What’s Feasible
You can’t boil the ocean. Pick one function within that value chain—one area you know intimately. This is where you’ll pilot your first experiment. To do this responsibly, you have to understand two things: what knowledge is required, and what cognitive actions are you asking the system to perform?
This is a perfect first task for an AI partner. Don’t do all the heavy lifting yourself.
“Get the AI to break it down first and then just critique it… that is way less work.”
Have the AI draft a map of the knowledge (Factual, Conceptual, Procedural) and the cognitive actions (Remember, Understand, Apply, Analyze, etc.). Your job is to critique and refine that map. This tells you if you’re asking the system to be a simple calculator or a creative partner, and it grounds your expectations in reality. Because if the foundational knowledge isn’t documented, you can’t automate the process.
“If you don’t have your processes documented today, you will need those processes before you can ask AI to follow the process. Right.”
4. Decompose to TOTE Cycles: Build a System for Quality
Once you have a specific task, you break it down into TOTE (Test-Operate-Test-Exit) loops. This isn’t about rigid, bureaucratic control; it’s about creating a rhythm of quality and building confidence through tight feedback. It’s the practical way we manage systems that are powerful but not perfect.
- Test (T1): What is our goal? How will we know we’ve passed?
- Operate (O): Take the action. Run the procedure or the AI model.
- Test (T2): Did we pass? Did we meet the goal?
- Exit (E): If yes, we’re done. If no, we loop back to Operate.
This is our quality control. It’s how we ensure we don’t build on a faulty foundation. When working with AI, this is non-negotiable.
“If it’s the wrong answer, don’t keep going on the back of the wrong answer. Go back to where it started being wrong.”
5. Build the Minimum Lovable Pilot (MLP)
Our goal is not a perfect, all-encompassing system. It’s a pilot that your team loves to use because it removes a frustration they feel every single day. The “Lovable” is more important than the “Minimum” or the “Pilot.”
“If your people won’t do it, it doesn’t matter how cool it is… the most valuable place for you to start is where the friction is felt by your team, because that’s actually going to motivate them to work on it.”
An MLP solves a real, felt pain point. It has a human in the loop, and it demonstrates its value quickly. This is how you earn trust, build momentum, and create a grassroots culture of innovation that starts to compound.
The Goal: A Compounding Advantage
This isn’t about a single win. It’s about building a series of small, meaningful advantages that stack up over time.
“If you can figure out how to integrate AI into your processes in a way that makes sense to you, you will get advantage and that advantage compounds. But if you can’t do that, you never start compounding the advantage.”
This grammar is the engine for that compounding. It’s a repeatable pattern that turns one win into a scalable capability. It’s how you move from just “doing AI” to building a truly AI-native business. And it starts with a simple commitment:
“If you leave this workshop and you don’t know what to do next, then I failed.”
This grammar is your “what to do next.”
RFQ Workflow — Application Note
- Primary grounding: read
Discovery Documentation/Safety Shower RFQ Workflow.mdfirst to map CF01–CF04 to your tasks. - CF01 Intake & Triage: Anchor hunt and completeness tests reduce friction by turning large RFQ zips into a verified working set.
- CF02 Requirements Analysis: Enforce repo‑relative citations with Doc_No + Section_Title + Page/Line_Range; pre‑draft RFIs for IP/EMC when offers lack explicit evidence.
- CF03 Solution & Deviations: Build per‑component hazardous‑area table; compute weakest‑link; run variant dependency (FLP vs non‑FLP) and power alignment checks; stage Evidence Pack items before decisions.
- CF04 Proposal & Handoff: Refresh annex manifests to enumerate only referenced items; mirror EP Item_IDs in CSV Decision_Notes; ensure all links resolve.
References
- Process:
Design & Build Workshop/MVP Documentation/Safety Shower MVP.md;Design & Build Workshop/MVP Documentation/MVP Sequencing.md - Standards:
Design & Build Workshop/MVP Documentation/Standard Work/reference/Citation_Style.md;.../templates/Spec_Crosswalk_Template.md;.../templates/Spec_Reference_Register_Template.md;.../reference/Evidence_Pack_Checklist.md