Growth engineering

Build the software behind acquisition and growth.

Choose the constraint: publishing, first-party analytics, attribution, or the operational software that turns demand into a repeatable process.

Find the blocked layer Show me the current loop

Remove the constraint that makes acquisition wasteful

Sometimes the constraint is discoverability. Sometimes traffic arrives but attribution is unusable. Sometimes the data exists and nobody can operate it. I work on the software and data behind those failures.

Start with the system closest to the visible failure.

Growth engineering loop from publishing and acquisition through routing, first-party measurement, decisions, and the next controlled release — The loop is only useful when every step leaves inspectable records. A ranking, click, or conversion is an outcome to measure—not a deliverable I can promise.

Choose the blocked layer

Start where the result breaks down

Each system explains the fit, working evidence, rough range, and next step for a real build.

Publishing / data / AIWorking system · controlled acquisition layer implemented

Programmatic publishing system

Turn keyword exports, source corpora, and media into reviewed, scheduled, multilingual static sites, then record indexation and controlled acquisition experiments per domain.

8 public renderer architectures · validated 50,000-page stream · 24 versioned acquisition techniques; 2 automated transports

Review this system

Analytics / search visibilityWorking multi-property self-hosted system

Private analytics and search visibility

Measure first-party traffic, campaigns, landing sources, search visibility, and crawler activity across multiple properties without cookies or long-lived visitor profiles.

Approximately 3 KB gzipped stateless tracker · 84-operation OpenAPI contract · permanent rollups with safe raw-data cleanup

Review this system

Attribution / routingWorking alpha · private builds only

Traffic routing and attribution engine

Route visits, preserve the complete decision path, accept conversions, and send callbacks without placing the tracking core inside somebody else’s SaaS account.

128 MB and 256 MB resource profiles · seven recorded hard-crash recovery scenarios

Review this system

Internal tools / operationsWorking food-operations app · private finance remains secondary evidence

Custom internal tools for operations

Custom internal tools for established operations that have outgrown business-critical spreadsheets and failed off-the-shelf software.

Working food-operations app: purchasing, FIFO stock, recipes, and order margin.

Review this system

First inspection

Match the visible problem to the evidence you need

The first useful change depends on the visible failure. More pages, events, dashboards, or automations can make the wrong system harder to repair.

Observed problem	Best starting point	Evidence required before expansion
The site cannot publish useful surfaces consistently.	Programmatic publishing	Source coverage, approval state, rendered routes, release validation, and crawl observations.
Traffic exists, but the source and landing context disappear.	Private analytics	First observed source, landing page, goal path, retention rules, and report reconciliation.
A visit must be routed and matched to a later conversion.	Traffic attribution	Click identity, compiled route, conversion idempotency, postback result, and recovery behavior.
The team sees the problem but cannot act without spreadsheets and manual joins.	Operations software	Named owner, repeatable workflow, decision inputs, failure states, and a recoverable source of truth.