Website:
a1techdeals.com
Job details:
PAY
Base: $3,000 USD/month.
MRR match: if your own product earns over $1,000/month verified, we add 1.5x every dollar of MRR above that first $1,000, up to $10,000 total comp. You're parking your product to work with us, so we price what you're giving up.
Worked examples: $500 MRR = $3,000. $2,000 MRR = $4,500. $4,000 MRR = $7,500. $5,700 MRR or above = $10,000.
The first month is a paid probationary evaluation at your figure. Pass it and the same base continues, with revenue share added on top: some of our services move revenue directly, our buy-box ranking and repricing engines among them, and your share is tied to the measurable impact of what you ship. The shape of your revenue share is set by how your probation month went.
Full-time contractor, fully remote.
HOURS AND COMMITMENT
Full UK working hours: 9am to 5pm UK time, Monday to Friday.
This role is your only engagement. No other job, no freelance clients, no studies. We pay for your full attention and we check.
ABOUT US
A1 Tech Deals is a UK electronics marketplace, and one of the fastest-growing e-commerce companies in the country. Our sales trajectory: roughly $2m in 2024, $20m in 2025, and we're on track to pass $100m USD annualised this year. That's 4.6x growth year on year, over 175,000 orders in the last twelve months, built with a team you can count on one hand.
You'd be employee-early. Not "early stage" in the sense of no revenue and a pitch deck, but early in the sense that the second engineer at a nine-figure-run-rate company gets: real scale, real problems, and a direct line between what you ship and what the business earns.
The stack: a Next.js storefront, two React and FastAPI dashboards, and about 40 backend services on GCP Cloud Run and Supabase. Merging to main is the production deploy.
Here's the part that matters: this company is run with fleets of AI agents. Scheduled agents sweep production logs and open fix PRs. Agents watch session replays and file UX issues, research products, assess held orders, draft seller emails. The founder operates multiple Claude sessions in parallel all day. We are not "exploring AI". It is how the work gets done here, and we're hiring the second person who works this way.
WHAT YOU'LL DO
Own features end to end from a short brief: design, build, test, ship, verify in production. Nobody reviews your code before merge. That's not a perk, it's the responsibility.
Work at the pace this setup makes possible. A feature that takes a traditional team a sprint ships here in days.
YOU
You've built and shipped your own product with real users and real revenue. We don't care if it made $100/month or $10k/month. We care that strangers paid for something you built, and that you kept it running.
You already work with AI agents daily as your primary way of building, not as autocomplete.
Curious by default: you try new tools and models the week they appear and keep what earns its place.
Strong fundamentals underneath: SQL you can reason about, TypeScript and Python, an eye for what breaks in production.
HOW WE HIRE
A short application, one conversation, then a paid work sample in our real repo using whatever tools you like. No LeetCode, no whiteboards, no rounds where AI is banned.
HOW TO APPLY
Start your application with the word "harness", then include:
- Proof of a product you built that had real users and real revenue: a TrustMRR profile, a Stripe dashboard screenshot, an app store listing with reviews, or anything equally concrete. Small numbers are fine. No product, no reply.
- A link to something you built with AI agents in your own time, not for an employer.
- A link to any public code, PRs, or projects.
- Your confirmation that you're available 9 to 5 UK time with no other commitments.
Applications missing any of these won't get a reply.
Click on Apply to know more.