Which parts of your job can AI already do?
Not someday. Today, with tools anyone can buy. An honest, task-by-task answer — plus what to learn next.
Job displacement doesn't announce itself. Nobody taps you on the shoulder and says "40% of your role got automated last quarter." It shows up later, sideways — hiring freezes, "we're restructuring," a job search that inexplicably takes a year.
The people who get hurt worst are the ones who never looked. Not because the answer was hidden, but because the question felt uncomfortable.
This tool makes the question take 30 seconds. Pick your role — or paste your actual weekly tasks — and get:
- A task-by-task breakdown: what AI does end-to-end today (🔴), what it turns into a copilot (🟡), and where your human moat is (🟢)
- An overall exposure score, calibrated to shipping tools, not lab demos
- What to learn next, in order of leverage
The point isn't fear. It's that automatable ≠ doomed — it means the shape of the job is changing, and the people who see the shape early get to choose their position in it.
1. Pre-analyzed roles (no signup, no key). Nine common roles — accountant, recruiter, support executive, content writer, junior developer, marketer, data analyst, sales exec, teacher — hand-calibrated against what deployed AI tools actually do in 2026.
2. Your exact job (bring your own API key). Paste your job title and real weekly tasks; Gemini analyzes them with the same calibration rubric. This is a static page with no backend — your key lives in your browser's localStorage and calls Google directly. Get a free key at aistudio.google.com/apikey; the free tier covers this comfortably.
git clone https://github.com/ramsai676/amiautomatable
cd amiautomatable
npm install
npm run dev # http://localhost:3000
npm test # scoring + role-library test suite
npm run build # static export to out/Stack: Next.js 14 (static export) · TypeScript · Tailwind · Gemini API
(gemini-2.5-flash, schema-enforced JSON responses, called straight from the
browser). No backend, no analytics, no tracking.
| Score | Meaning |
|---|---|
| 85–100 | Shipping tools do this end-to-end with light review |
| 60–84 | AI does most of it; a human directs and finishes |
| 35–59 | Real speed-up, but human judgment carries the outcome |
| 0–34 | Fundamentally human: trust, presence, accountability, relationships |
Overall exposure = mean across tasks. The custom analysis prompts Gemini with
this exact rubric and constrains the response with a responseSchema, then
re-validates client-side — so results are structurally guaranteed.
Five problems people have but haven't noticed, five open-source tools, five days. AmIAutomatable is 2 of 5.
MIT — fork it, rebrand it, show it to your team before your boss does.