Front-facing portrait of David Akilo, AI systems engineer, standing with arms crossed
AI Systems Engineer: D2C Ecommerce

I build the automations your ops team wishes they had.

Custom Claude Code systems and AI workflows for founder-led ecommerce brands: translation pipelines, BI dashboards, lead capture, competitor intel. Shipped and running in production, not slideware.

3production systems shipped
2ecommerce clients, verified
// The Offer

What happens when you click that button

A focused engagement: I map your manual workflows, flag what's automatable, and ship the first working system inside the sprint. Not a 40-page deck.

Book an Automation Audit
// Verified in production
3
retail stores kept in sync across a Stockholm agency's client base
4
ad, Shopify and email data feeds unified into one live dashboard (personal build)
3
tracking integrations pre-wired for Gotain: Meta Pixel, GTM, GA4
10+
content hooks generated per cycle in MileFlying's planning tool
// Stack

Your Stack, Automated

I build directly into your existing ecosystem using industry-leading infrastructure and tools.

Claude Code Klaviyo Meta Ads Next.js Shopify Partners Supabase Vercel n8n
// Services

Four systems. Zero fluff.

I don't sell "AI strategy." I write and ship the workflow, wire it into your stack, and hand you something that runs without me in the room.

Audit sprints from $150

01

Multi-store sync systems

Translation and publishing pipelines that keep product content, pricing, and copy consistent across every storefront, no more manual re-uploads per region.

02

Custom BI dashboards

One source of truth for revenue, ad spend, and channel performance, replacing the Frankenstein spreadsheet your team dreads updating every Monday.

03

AI lead capture

Branded, tracked capture forms that plug directly into your ad and analytics stack, so a lead doesn't sit in an inbox untracked.

04

Competitor intel & content planning

Automated tracking of competitor moves paired with a hook-generation engine, built to keep a content calendar full without a strategist glued to spreadsheets.

All of the above are built as Claude Code agents wired directly into your stack, not a chatbot bolted onto the front end.

// Verified work

Real systems, real clients.

Every project below is live or was live in production. No composite case studies, no "results may vary."

Multi-store sync Translation pipeline

Keeping Skultuna in sync across three storefronts

Built via a Stockholm agency partner: a multi-market translation and publishing sync for Skultuna, a Shopify Plus retailer, keeping roughly 800 SKUs consistent across three stores, US, Norway, and a dedicated B2B storefront. One publish action now propagates translated product content everywhere, so an update in one market lands correctly in every other store, automatically.

  • Eliminated manual re-entry of translated product copy across all 3 stores
  • Single publish action propagates to US, Norway, and the B2B storefront
  • Version-consistent copy across roughly 800 SKUs, zero drift
via agency partner · Skultuna
Screen recording of the Skultuna publishing and translation manager syncing product content across storefronts
BI dashboard Reporting Personal build · demonstration data

Pulse's multi-channel reporting, without the spreadsheet

Killed the manual multi-channel spreadsheet pulled together by hand from ad platforms, Shopify, and email. Replaced it with a single dashboard that refreshes itself, so leadership sees live numbers instead of waiting on a report each Monday.

  • Consolidated four data sources into one live report
  • Removed a recurring weekly manual reporting task entirely
  • Built on the client's existing stack, no new platform to learn
Pulse · personal build
Screen recording of the Pulse BI dashboard showing unified revenue, ad spend, and order data using demonstration data
Lead capture Tracking

A tracked lead-capture form for Gotain, built and ready to ship

A branded lead-capture form matched to Gotain's site, with Meta Pixel, Google Tag Manager, and GA4 tracking pre-wired out of the box. Built as a self-initiated prototype and reusable tool, an estimate was sent for the engagement but never confirmed, so this is shown here as a completed build, not a live client deployment.

  • Branded capture form matched to the client's site
  • Meta Pixel, GTM, and GA4 tracking pre-wired
  • Built as a reusable tool, ready to deploy
Gotain · prototype build
Competitor intel Content planning

MileFlying's content calendar & hook generator

A planning tool that tracks competitor content and status across formats: Studio, Cell Phone, Green Screen, Carousel, and generates a bank of tested hook angles per post, so the calendar never runs dry.

  • 10-hook bank generated per content slot, ranked by pattern
  • Full month view with format and upload status at a glance
  • One-click status tracking from upcoming to uploaded
MileFlying · content operations
MileFlying content calendar interface showing April 2026 schedule and a panel of ten AI-generated content hooks for an Amex Cobalt Instagram Reel
David Akilo smiling, wearing glasses and a dark green polo shirt, holding a black coffee mug
// About

I write the code myself. No agency layer, no account manager relay.

I'm David Akilo, an AI systems engineer who builds Claude Code automations for D2C ecommerce brands that are tired of manual workflows eating their week. No decks, no "AI transformation roadmaps." I sit with your actual stack, find the repetitive task costing you hours, and ship a working system for it.

Most projects on this site are systems I built and shipped for a real operator: a Stockholm agency's retail client, a media planning team. A couple are self-initiated builds and prototypes, clearly labeled as such where they appear.

Claude Code automations Ecommerce integrations Production-first builds
David Akilo
AI Systems Engineer
// FAQ

Questions before you book.

The short answers, no sales-deck spin.

Does this replace our existing tools or developers?

No. Every system I build works alongside what you already have. It's wired directly into your existing stack, Shopify, your ad platforms, your dev team's tools, not a replacement for a dev team or a rip-and-replace of tools you already rely on.

How fast do projects typically ship?

An Audit & Build Sprint ships the first working system inside the sprint itself, that's the point of it. Larger, multi-system builds take longer and get scoped per project once we know exactly what needs to be automated.

Are ongoing or retainer engagements available?

Yes. Past the initial sprint, ongoing and retainer engagements are available for teams who want a standing engineer on their automation systems rather than a one-off build.

What tech stacks do you support?

I build primarily into Shopify, Klaviyo, Meta Ads, and n8n, with systems hosted on Vercel and Supabase, and Claude Code as the automation engine underneath. If your stack looks different, tell me what it is, the point is wiring into what you already run, not migrating you to something new.

What happens after the audit sprint ships?

You own what's built, it keeps running without me in the room. From there, most clients either scope a second system, or move to an ongoing retainer if there's a steady pipeline of automation work.

// Contact

Let's find the bottleneck eating up your team's week.

Book an Automation Audit, a straight look at your current workflows and where a custom system pays for itself fastest. No sales deck, no retainer pressure.

No sales pitches. You talk directly to the engineer who writes the code. Response within 24 hours.