Senior Full-Stack Developer — Python & Next.js | Part-Time Growing to Full-Time & Technical Lead | AI Automation & Clien

We're ScaleSage — UK-based AI automation and client acquisition specialists, one venture inside The Machine Collective. Our platform is nearly built: Python/FastAPI back end, Next.js front end on Vercel, Postgres on Supabase, Stripe, cold email through the Instantly and Apollo APIs, a conversational diagnostic funnel, a partner commission engine, CI-gated deployments to a Mac Mini production box behind Cloudflare Tunnels, and a growing set of AI agents.

You'll take the platform over from the current developer during a structured handover, finish it to a proven standard, and then build what comes next — a client platform, a content production product, and a search and AI visibility product — all scoped and all yours to own.

Here's the honest situation. A lot has been built. From now on nothing is called done until it's demonstrated working, estimates are real and kept, and every hour is accounted for.

If that sounds like pressure, this isn't your role. If it sounds like how good engineering should work anyway, keep reading.

HOW THIS WORKS

We start on a monthly retainer at ~20 hours a week, on a fixed weekly pattern you propose and we agree, with at least three hours of daily overlap with Bangkok and Manila working hours.

Nothing here is promised. Everything is earned. The path is written into the contract: part-time to prove the fit, expanded part-time as you deliver, full-time when the business and your delivery earn it, and then technical lead across the group's ventures. Full-time follows delivery, not time served.

What we hire for, in order: finishing what you start; proving it works before you call it done; clear, precise written and spoken English, because nothing can get lost in translation here; being extremely organised; and an eye for detail, for design, and for what premium quality actually is. Our clients pay for premium and can tell the difference.

WHAT YOU'LL INHERIT

FastAPI services on Docker Compose on a Mac Mini behind Cloudflare Tunnels and Tailscale; Postgres on Supabase with Alembic migrations and row-level security; GitHub Actions CI gating deploys; Instantly, Apollo, Hunter, Google Places and Companies House APIs; Stripe with webhooks; HeyGen and ElevenLabs for generated video and voice; an e-signature flow; scheduled jobs; LLM agents with per-run cost tracking; Slack alerting. All of it real, all of it in production today.

WHAT YOU'LL ACTUALLY DO

The 30-day trial starts on day one and runs alongside the outgoing developer for as long as he's there, overlapping for the remainder of his notice period so you inherit the platform safely before he's gone.

Own the platform end to end. Nothing is called done below proven, with evidence:

Daily written EOD: completed, active, blocked, next — with hours
Four status words for all work — architected, coded, live, proven
One live demonstration call a week, working things shown on screen, plus short recorded walkthroughs
Blockers raised the same day they appear
Time tracking through an agreed tool
Any other client work disclosed in writing before it starts

Post-trial. You own the platform and scope and build what's next: a client platform, a content production product, and a search and AI visibility product. Not a phased roadmap — you're shipping what the business needs as it needs it.

A candidate who takes clean ownership of the platform and closes out the handover meets the bar. One who backs every claim with a proven demo is doing this role well. One who also spots what's inefficient or under-specified and improves the plan is exceptional.

MUST HAVE

Python and FastAPI in production
Postgres and Supabase, schema design and migrations, background jobs and schedulers, webhooks
Next.js and React on Vercel, Tailwind, responsive layouts that look right on a phone, CSS animation and motion, and the ability to take a Figma or visual reference and build it to the pixel
Stripe subscriptions and one-off payments, REST API integration with third-party SaaS

Email sending and deliverability
Docker, CI pipelines, Linux, Cloudflare Tunnels, backups and restore, secrets handled properly
LLM APIs, prompt and agent design, evaluation of outputs
Automated tests, QA on real devices, security and data-protection basics
Documentation good enough that someone else could take over
Excellent written and spoken English — you'll write daily EODs and speak on a weekly demo call, and English is tested live on the screening call
Self-managing: able to take a brief and ship without hand-holding
Reliable internet, a quiet workspace

If you're strong in most of these and honest about the rest, apply.

NICE TO HAVE

Experience taking over an existing, undocumented codebase and making it safe to change
Background in an agency or client-services environment where "done" has to be provable to someone else
Exposure to content production tooling or SEO/AI-search-visibility work
Working knowledge of UK GDPR and PECR as they apply to cold email: suppression, unsubscribe handling, eligibility
Active GitHub with public commits in the last 90 days
Entrepreneurial mindset — you've tried to build something yourself

You don't need to know everything already. But you need to be the kind of person who can learn a new tool fast, build with it, and document what you did so the rest of the team can use it.

HOW TO APPLY

Send:

A short recorded video, under three minutes, on why this role and how you'd approach the first week.
Answers to the seven questions below, directly in your application. Applications without them are not read. Give real examples — generic answers are exactly what this is screening against:
Describe a system you built end to end and are proud of: the stack, what it does, what's live today, and what you'd do differently.
Have you taken over someone else's system? Give an example: what you were handed, what was missing, and how you got to the point of being able to change it safely.
Give an example of a deadline you missed. When did you tell the client, what did you say, and what changed afterwards?
Describe something you shipped that later turned out not to work in production. How was it found, and what did you change so it wouldn't happen again?
What is your current client load in hours per week, and what fixed windows could you give this role, in Bangkok time?
Send a link to something you've built where the front end is yours, and tell us one thing about it you'd fix if you had an hour.
You'll have Claude Code and Codex. Tell us how you use them and how you know what they produce is right.

If your answer to question 4 is "that has never happened to me," this isn't the role for you.

In return: real freedom in how and when you work, a direct line to the founder, a documented system, and a company that wants you to grow into leading it.

Process

Written screen → 30-minute call → short technical test → two client references → contract with a 30-day trial.