WORK WITH MEOpen for new builds · 2026
← FDO.CODES

BLOG

(Notes on building with AI)
FDO.CODES Field Notes 38: EXTRACT. CHECK. REVIEW. A monochrome invoice passes through three architectural gates toward a lime exception tray.Use an invoice harness for intake, with payment kept separateExtract documents into a review queue, check the numbers and surface exceptions before accounting decisions.Read note FDO.CODES Field Notes 37: AGREED? OR ASSUMED? Two colleagues in conversation in black-and-white, a fine-line waveform and lime review marker.Separate agreed actions from suggestions in meeting notesA meeting harness should preserve uncertainty, resolve ownership and prevent duplicate tasks.Read note FDO.CODES Field Notes 36: ONE BRIEF. CLEAR LIMITS. One source page feeds three publication cards through a lime validation gate.Turn one approved brief into content without inventing the storyUse a structured brief, a source list and explicit claim checks to generate reusable channel drafts.Read note FDO.CODES Field Notes 35: ANSWER WITH EVIDENCE. Monochrome customer support professional wearing a headset, a source document linked to a lime reviewed-answer card.Build a support-answer harness that knows when to stopRetrieve approved knowledge, draft within policy and route unsupported answers to a person.Read note FDO.CODES Field Notes 34: COMPARE THE SAME THING. Monochrome quote stacks aligned to a comparison grid, one lime missing-data cell.Compare supplier quotes with a reusable evidence tableFix the comparison fields, preserve the source evidence and flag missing terms before a person chooses a supplier.Read note FDO.CODES Field Notes 33: LEAVE A USEFUL HANDOFF. Two fictional developers exchange a document in black-and-white editorial photography with a lime bookmark.End an AI coding session with a handoff, not a memory dumpA concise record of state, evidence and open work helps the next session resume without trusting a stale summary.Read note FDO.CODES Field Notes 32: EARN THE BIGGER MODEL. Two precise routes through geometric gates end at one lime acceptance node.Make a bigger coding model earn its placeUse the same acceptance checks to compare model choices, including retries and human review.Read note FDO.CODES Field Notes 31: SHIP ONE THIN SLICE. Isometric architecture cut into layers with one connected lime slice through the full structure.Ask AI to build one working slice before the whole featureConnect the interface, logic and storage for one realistic path before multiplying screens and features.Read note FDO.CODES Field Notes 30: SMALLER CONTEXT. Black-and-white developer portrait with three selected document cards and a lime edge.Build a small context pack instead of pasting the whole repositoryGive the agent the files, constraints and examples that define this change, then let it retrieve more when needed.Read note FDO.CODES Field Notes 29: REPRODUCE. THEN REPAIR. A broken circuit path becomes a lime closed loop on a near-black technical drawing.Give your coding agent a failing example before asking for a fixA reproducible failure turns a vague debugging request into a change you can evaluate.Read note FDO.CODES Field Notes 28: TOMORROW, PREPARED. Black-and-white editorial photograph of a fictional professional at a breakfast table with notebook, a minimal next-day timeline with one lime preparation block; warm-white canvas.Build a tomorrow brief that leaves your calendar aloneUse a read-only calendar and your own notes to prepare for the day without giving an agent control of it.Read note FDO.CODES Field Notes 27: WHAT CHANGED? Near-black background, warm-white typographic headline, architectural fine-line layers of documents with one lime highlighted difference; no people.Create a digest that tells you what changedWatch a few trusted sources and send yourself only meaningful changes, with links back to the evidence.Read note FDO.CODES Field Notes 26: SCREENSHOT TO ACTION. Black-and-white editorial portrait of a fictional man looking at his phone, beside a stack of monochrome screenshot cards sorting into three minimal lanes; one lime destination.Give your screenshots a next actionA small review queue can turn saved screenshots into useful reminders, references and tasks.Read note FDO.CODES Field Notes 25: KEEP THE RECEIPT. Minimal black-and-white still life of a paper receipt, shoe box and calendar; one lime reminder tab, precise thin-line annotations without extra words.Build a return-window reminder from the receipts you already saveUse AI to organise purchase evidence while letting verified dates and ordinary rules drive reminders.Read note FDO.CODES Field Notes 24: SAVE THE DECISION. Black-and-white editorial portrait of a fictional woman speaking into her phone; a fine-line waveform leads to one lime action card.Turn a voice note into a decision you can actually useCapture an idea once, then turn it into a small, reviewable action without inventing the missing details.Read note FDO.CODES Field Notes 14: Give AI an owner. Three architecture blocks connect through a lime bridge on warm white.Who Owns the AI Between Demo and Delivery?The AI Solutions Architect connects business decisions, system boundaries and operational ownership so useful prototypes have a credible route into daily work.Read note FDO.CODES Field Notes 15: Budget for change. Five geometric planning blocks and a flexible line on black.Build a Five-Year AI Budget That Can ChangePlan for durable capabilities, fund delivery in stages and revisit assumptions as the economics and use cases change.Read note FDO.CODES Field Notes 16: Rethink the work. An angular process path and a shorter lime route on warm white.Revisit Your Strategy Where the Work Has ChangedUse AI capability changes to revisit specific business assumptions, with evidence from the workflow and a clear route to adoption.Read note FDO.CODES Field Notes 17: Train the decision. Three choice cards show a question mark, magnifying lens and check mark on black.AI Literacy Starts With Everyday DecisionsPractical training helps people decide what to share, what to verify, when to act and when to stop.Read note FDO.CODES Field Notes 18: Bound the access. An AI node sits within a geometric enclosure with one lime doorway on warm white.Secure AI Begins With Bounded AccessApply familiar security disciplines to the new pathways created by models, tools and retrieved content.Read note FDO.CODES Field Notes 19: Context has boundaries. A document passes through a lime aperture while other sheets remain outside, on black.Decide What Data Your AI Is Allowed to SeeAn approved model does not automatically make every document appropriate input. Govern the complete information path.Read note FDO.CODES Field Notes 20: Policy into practice. A document becomes three control gates along a path, with a lime accent on warm white.Put Company Policy Inside the Agent’s HarnessTranslate written rules into permissions, approval gates, evidence and exception handling that operate with the workflow.Read note FDO.CODES Field Notes 21: Make review meaningful. A magnifying lens examines a decision node beside a lime pause symbol on black.Human Oversight Needs Time, Context and AuthorityA review step is useful when the person can understand the evidence, challenge the recommendation and change what happens next.Read note FDO.CODES Field Notes 22: Rehearse the response. A circular operational loop has a lime stop switch and a recovery path on warm white.AI Incidents Need a Rehearsed ResponsePlan how to contain an agent, preserve useful evidence, restore service and learn from mistakes before an incident occurs.Read note FDO.CODES Field Notes 23: Govern the outcome. Four minimalist measurement dials with one lime highlight on black.A Governance Scorecard for the AI Leadership TeamUse a compact view of value, reliability, control and adoption to decide which AI services to expand, improve or retire.Read note A horse wearing a lime harness on a warm-white background, titled Power needs a harness.The model is the stallion. Build the harness.Power becomes useful when the surrounding system gives it direction, boundaries and feedback.Read note A client desk and connected workflow on a black background, titled Start at the desk.Start at the client’s desk, before opening the model catalogueAn embedded approach finds the real work, its exceptions and the people who must trust the result.Read note European storage and inference illustrated as separate connected systems, titled EU storage. EU inference. Ask both.EU data residency and EU inference are different questionsA European storage location is one part of the architecture, not a complete answer about processing.Read note A branded cost-per-success illustration on a black background, titled Measure cost per success.The real AI bill is cost per successful workflowToken prices matter. Retries, review effort and failed work determine whether the system is economical.Read note A bounded agentic workflow on a warm-white background, titled Narrow job. Clear boundaries.A useful agent needs a narrow job and a clear action boundaryAn illustrative onboarding pipeline shows where model judgement helps—and where ordinary software should decide.Read note A branded evaluation illustration on a black background, titled Test the outcome.Reliability means passing the same standard, not repeating the same sentenceEvaluate outcomes, exceptions and actions across repeated runs before expanding autonomy.Read note Code and a delivered application connected through a verification step, titled Code is one step.Your coding agent needs a delivery systemRepository context, bounded tasks and release evidence matter as much as code generation speed.Read note Three specialised agents feeding a shared result, titled Every agent needs a job.More agents should earn their placeParallelism is useful when the work separates cleanly and the coordination cost is justified.Read note A prototype progressing from sketch to working interface, titled Build to learn.Build the prototype around the question you need answeredAgentic development can accelerate learning when the product slice and production gap are explicit.Read note An iterative improvement cycle around software components, titled Design it to evolve.The harness must evolve with the model and the businessKeep interfaces stable, measure changes and retire controls that no longer help.Read note FDO.CODES Field Notes 01: Agents earn their place. The agent loop of model, tools and check, with an exit once the goal is met.Agentic AI without the hype: where agents earn their placeAn agent is a language model, a set of tools and a loop. Where that combination pays off in real organisations, and where a plain workflow is still the better choice.Read note FDO.CODES Field Notes 02: Fit beats the leaderboard. A cost-versus-quality chart where the model that fits your task is not the top-ranked one.Choosing a language model in 2026: what actually mattersNew models arrive every few weeks. The leaderboard rarely decides what works in production; your tasks, your data and your constraints do.Read note FDO.CODES Field Notes 03: Frame, orchestrate, evaluate, ship. Four checkpoints from idea to production, the last one lit.Frame, orchestrate, evaluate, ship: how I take AI from idea to productionThe four stages I use to turn an AI idea into a system people rely on, and the question each stage has to answer before the next one starts.Read note
(Contact)

LET'S
BUILD.