Use case

A recruiting pipeline shaped to how you actually hire.

Generic ATS tools model a generic funnel. Your hiring isn't generic: your stages, your questions, the things you actually care about in a CV. A generated pipeline tracks candidates through your process, uses AI to pre-screen against your criteria - with a human on every decision - and never loses a candidate in an inbox.

What your generated system includes

Data model

  • Candidates: role, source, stage, CV, notes
  • Roles with their stage sequences and criteria
  • Interviews with structured feedback forms
  • Talent pool of past candidates worth keeping

The flow

  • Applications arrive by form or forwarded email
  • AI screens the CV against the role's criteria - suggestion only
  • You review the screen and advance or decline
  • Each stage assigns its tasks: schedule, brief the panel, send the exercise

Automations

  • Acknowledge every application immediately
  • Nudge when a candidate sits in a stage past your SLA
  • Decline emails - drafted by AI, sent only on your approval
  • Weekly pipeline summary per open role

The human rule

  • AI never rejects anyone - it annotates, you decide
  • Every AI screen is visible and overridable
  • Consequential-decision review is built into the flow
  • Full audit trail per candidate

Try this exact prompt

Paste into Chromoly

Build a recruiting pipeline. Candidates apply via a form with CV upload. Stages: Applied, Screened, Interview 1, Exercise, Final, Offer, Hired/Declined. AI screens each CV against the role's criteria and writes a short annotation for me - I make every advance/decline decision. Acknowledge applications automatically, nudge me if a candidate sits anywhere over 5 days, and draft decline emails for my approval.

Hiring decisions affect real people - the system is deliberately structured so AI annotates and humans decide, with every decision logged.

Why AI-assisted screening needs a human in the loop

Fully automated CV rejection is where AI hiring tools earn their bad reputation - biased proxies, silent errors, no accountability. The generated pipeline takes the other bet: AI reads and annotates (saving the hour of skimming), and a person makes every call, with the AI's reasoning visible and overridable. That's not a compliance fig leaf; it's the same architecture principle the whole platform runs on - judgment at the edges, humans on consequences.

Frequently asked questions

Is AI screening of candidates even okay?

Structured the right way, yes: AI as an annotator with a human making every decision, reasoning visible, decisions logged. Chromoly will not build fully automated rejection flows - consequential decisions about people keep a human in the loop by design.

Can candidates see anything?

If you want: an application-status view is a small addition, and acknowledgment plus decision emails are automated (drafted for your approval). Many teams keep candidate-facing surfaces minimal and use the system internally.

Who maintains the system after it's built?

Chromoly does - monitoring, security patches, AI model upgrades, and integration migrations are platform-owned for the lifetime of the account. Changes you request run through a dependency-aware preview showing the blast radius before anything ships.

Can I import my existing data?

Yes - export your current spreadsheet or tool to CSV and import with column mapping, or attach the spreadsheet to your prompt and Chromoly shapes the data model from it directly.

Never lose a good candidate to an inbox again

Describe your process in plain language. Preview the working system free - deploy when it's right.

Start free →

150 credits/month free forever · EU-hosted · maintained for life