LUTHFI SEPTIAN — AI GENERALIST · AI TRAINING, DATA ANNOTATION & EVALUATION
Instead of prompting one chatbot and hoping for the best, I built a working system: four focused AI offices — engineering, games, video, and marketing — each with its own pipeline, its own standards, and one person accountable for quality at the end of every line: me. The system turns AI assistance into finished, quality-checked deliverables.
One person can only wear so many hats well. So I split the work the way a real company would: four offices, each with a single specialty, a defined pipeline, and its own graduation standard. An office only earns the title “Graduated — Ready to Work” when all five of its projects have been built, audited, and passed through my quality gate.
The office that builds practical tools for client work — dashboards, calculators, workflow boards, and document generators. If a tool comes out of NewCode, it exists to make a real job faster or clearer.
The office that builds playable browser games — and where my QA changes shape: I review as a playtester, not an inspector. A game that works correctly but is boring fails Forge's quality gate.
The office that runs short-form video production as a pipeline — from a raw idea to a publish-ready package, with a watch-through QA step before anything is allowed out the door.
The office that handles client outreach as a funnel, not a blast — one personal, honest message per target, a maximum of three follow-ups, and an immediate stop on any rejection. No spam, no invented claims, ever.
The whole program — all four offices, all twenty projects — was designed, built, and graduated in one push.
Every number on this page is real. If a capability is partial, the page says partial. That rule is part of the system — it is the same standard I apply to client data.
Every project below was briefed by me, built to my specification, audited, and passed through my QA before it counted as done. Filter by office:
Showing all 20 projects
Every project in every office runs the same five-step workflow. The AI does the production; I do the directing and the judging.
I define the job: what it must do, who it serves, and what “good” looks like.
Structure, layout, and rules are settled before anything gets built.
The office produces the work against the brief — tools, games, scripts, messages.
Everything is tested by actually using it. Defects go back for fixing, not into the report.
Only after QA passes does a project count as done — and get recorded as done.
I am the owner, system designer, and director of all four offices. I write every brief, set every standard, and decide what gets built next. The offices are staffed by AI; the judgment is staffed by me.
Most importantly, I am the final quality gate. Each of the 20 projects above only graduated after an audit and my personal approval — the same way I review AI responses and annotated data for clients.
The system's statistics are never rounded up. When a capability is partial, it is labelled partial; when something is a template generator rather than live AI, it says so.
That is not a limitation of this portfolio — it is the product. Clients hire me to judge AI output and data accurately. A portfolio that exaggerates would fail its own QA.
Nothing ships without passing my QA. That includes this page.
The office system is proof of method. The services below are what that method is for — the work I do for clients and platforms.
Capabilities proven by the 20 projects above
Core services — AI training & data work since August 2026
If you need AI responses evaluated, data annotated, or Indonesian content reviewed by someone who treats quality as a gate — not a slogan — I have 40+ hours a week available and a system behind me that makes the work fast without making it sloppy. Send a brief; you'll get a straight answer on fit, timeline, and price.
Rates from US$15/hour — negotiable by volume.