A Recruiting Pipeline You Can See at a Glance
Turn scattered candidate updates into one clear pipeline view — who's where, what's stuck, and what needs a decision this week — instead of a spreadsheet nobody trusts.
A pipeline you can't see at a glance is a pipeline where good candidates quietly go cold.
Hiring stalls in the gaps — the candidate waiting five days for a yes, the role nobody's sourced for in a week. Store your stages and what 'stuck' means, and a scattered set of updates becomes one view that surfaces exactly what needs you now.
The best candidate in your funnel is also interviewing elsewhere. Every day they sit in 'awaiting decision' is a day you might lose them to a faster team.
Candidates don't drop out because they're rejected. They drop out because they're ignored while you decide.
The view is for action, not admiration. Its job is to put the two or three things that need a decision this week in front of you.
| Complexity | Easy |
| Tools needed | Claude Pro or above, Claude Desktop → Cowork mode, on macOS or Windows, Claude Projects |
| Time to build | ~20 min first time · ~5 min per update after that |
| Best for | Founder or HR lead running hiring across several open roles — Small companies hiring without a dedicated recruiter |
What this solves
Candidate status scattered across threads, notes, and memory — so the person who's waited a week on a decision is invisible until they withdraw.
The problem
You’re hiring for four roles and the status of each lives in a different place — an email thread here, a note from the hiring manager there, your own memory of who you owe a reply to. When someone asks “how’s hiring going?” you can answer vaguely, but you can’t quickly say which candidate has been waiting on a decision for a week, or which role hasn’t had a new applicant in ten days. So a strong candidate cools while you’re heads-down on another req, and you find out when they email to withdraw. The information exists; it’s just never assembled into a view that shows you where to act.
The fix is turning the scattered updates into one glance-able pipeline, refreshed whenever you need it.
Ingredients
- Claude Subscriptions — Cowork and Projects aren't available on the free plan
- Platforms & Modes — Cowork runs in the desktop app only — not on web or mobile. This recipe consolidates scattered updates into a status view — multi-step work that runs in Cowork
- Claude Projects — Your stages, your roles, and what counts as "stuck" stay the same. A Project stores them so every status is built the same way and flags the same risks
How it works
You track hiring continuously, and your stages and definitions of “stuck” are constant. A Project stores them so every pipeline view is built consistently and surfaces the same risks each time.
- Open Claude Desktop and click Cowork in the mode selector across the top (Chat · Cowork · Code).
- In the left panel, find Projects and click the + button.
- Choose Start from scratch. Name the project “Recruiting Pipeline” and let Claude set up its folder.
- You’ll know it worked when the project appears in the left panel with its own folder and an instructions field.
Your project has a folder on your computer — that’s where the two working files live.
- Ask Claude, right in the project: “Create two files in this project’s folder:
pipeline-rules.mdandcandidates.md. Leave them empty — I’ll fill them in.” (Or create them yourself in any text editor and save them into the project folder.) - Fill in both files using the descriptions below.
- Confirm Claude can see them: ask “List the files you can see in this project.” Both filenames should come back. If they don’t, see If It Doesn’t Work.
pipeline-rules.md
Your stages, what advances a candidate, and your thresholds for “stalled” — so the view flags risk by your standard, not a generic one.
Example: “Stages: applied → screen → onsite → debrief → offer. Stalled = no movement in 5 business days. Awaiting-decision over 3 days is a red flag. Roles with no new applicant in 7 days need sourcing attention.”
candidates.md
Your current candidates with stage, last activity, and any notes — updated as you go, or pulled live if your ATS is connected.
Example: “Role: AE. Priya — onsite done Mon, awaiting debrief (3 days). Marcus — screen scheduled Thu. Role: Support Lead — 0 new applicants this week.”
Open your Project in Claude Cowork. Give Claude the specifics for this run, ask for the main output, then follow up for any additional pieces you want. The exact wording for each prompt — with what it’s asking for and why — is in What you actually type below.
Check three things before you act on the view:
- At-risk items are at the top. The view’s job is to surface what’s slipping, not to list everyone neatly. If a candidate who’s waited a week is buried mid-list, the ranking failed. The stalled and awaiting-decision items should lead.
- “Stuck” matches your rules. Confirm Claude flagged delays by your thresholds from
pipeline-rules.md, not an arbitrary sense of slow. A flag you don’t agree with is a flag you’ll ignore. - The action list is small and specific. “Follow up with candidates” is not actionable. “Get a debrief decision on Priya (4 days), nudge sourcing on Support Lead” is. If the to-dos are vague, ask for the specific next move per item.
Before you act: does the view match what you know that isn’t written down? If a candidate told you verbally they’re pausing, the data won’t show it. Your memory plus the view beats either alone.
What you actually type
Name your files explicitly the first few runs, and ask Claude to show its work on anything that matters.
“Using `pipeline-rules.md` and the current candidate data, give me a pipeline status: candidates by role and stage, anything stalled past our thresholds, who's awaiting a decision from us and for how long, and the two or three things that need my attention this week. Put the at-risk items first.”
“Draft a nudge to the hiring manager for the two candidates stuck in debrief.”
A pipeline status with the at-risk items first: candidates by role and stage, anything stalled past your thresholds, who's awaiting a decision from you and for how long, and the two or three things that need your attention this week. Every flag traces to a threshold in `pipeline-rules.md`, not a generic sense of slow. If a candidate who's waited a week is buried mid-list, or the action list says 'follow up with candidates' instead of naming the next move per person, it missed — ask for the ranked version.
- It can't see updates you don't give it. Without an ATS connector, the view is only as current as your last edit to `candidates.md`.
- It won't capture what's verbal. A candidate who told you in person they're pausing looks fine in the data — your memory still has to check the view.
- It surfaces decisions; it doesn't make them. The debrief still has to happen — the view just stops it from happening a week late.
If it doesn’t work
- No Cowork tab in Claude Desktop — update the app to the latest version and confirm you’re on a paid plan; Cowork isn’t on the free tier. On Windows, Cowork also needs the Virtual Machine Platform feature enabled — if the tab still won’t appear, that’s the fix.
- Claude can’t see
pipeline-rules.mdorcandidates.md— the files aren’t in the project’s folder, or they’re in a different folder than the one the project owns. Open the project, check which folder it points to, and move the files there. Then re-run “list the files you can see.” - Stalled candidates are buried mid-list — the view is organized by role or stage instead of by risk. Ask: “Put anything stalled or awaiting a decision at the top, ranked by days waiting.” Then add that ordering rule to
pipeline-rules.mdso every future view leads with what’s slipping. - The flags don’t match your thresholds — Claude is using its own sense of “slow” instead of your rules. Name the file in the prompt — “flag delays per
pipeline-rules.md” — and check that the file states thresholds as numbers (“stalled = no movement in 5 business days”), not adjectives. A flag you disagree with is a flag you’ll learn to ignore.
Extra credit
Small additions that pay back the next time you run it.
- ATS connector — connect your applicant tracking system so the pipeline pulls live instead of you pasting updates. See the Connectors guide.
- Weekly auto-status — set a scheduled task to generate the pipeline view every Monday morning so the at-risk list is waiting for you.
- Live board — turn it into a live artifact the hiring team can open any time to see current status without asking you.
“A hiring process loses more good people to silence than to no.”
What this teaches you about Claude Cowork
The recipe is one application. The principles apply to everything you’d hand to Claude.
Consolidation surfaces risk. The danger in hiring isn’t a bad decision; it’s no decision while a candidate waits. Pulling scattered updates into one ranked view makes the slipping items impossible to miss — which is the whole point.
Your rules define the alerts. Storing what “stalled” means in your context is what makes the flags trustworthy. A generic tracker nags about everything; a view built on your thresholds tells you the specific things that actually need you.
A view for action, refreshed on demand. The recipe wins by being fast to regenerate, so the pipeline is current whenever you look — not a spreadsheet you update monthly and stop trusting. Pair it with a schedule and the at-risk list comes to you.
Who this is for
Founder or HR lead running hiring across several open roles in Small companies hiring without a dedicated recruiter (10–100 employees).
The pain: Good candidates going cold in the gaps nobody is watching
The outcome: A glance-able at-risk list that surfaces stuck candidates before they withdraw
Published June 23, 2026 · 0 views