How to run an interview loop as Chief of Staff and Founder's Office

People & HRFor Chief of Staff and Founder's Office3 apps12 steps~24 min to set up

Running an interview loop at 150 people means you're coordinating across six hiring managers, three recruiters, and a CEO who agrees to final-round interviews and then triple-books herself. The scheduling alone eats your Tuesday: you're manually cross-referencing Google Calendar availability, sending Calendly links that don't account for panel conflicts, copy-pasting interview feedback from Slack DMs into a Notion table, and chasing people down two days after the debrief to actually submit their scores. Nothing is broken enough to fix — it's just slow and manual every single time, and the candidate experience suffers because you're the bottleneck.

People & HRFor Chief of Staff and Founder's Office3 apps12 steps~24 min to set up
Outcome

What you'll set up

A calendar-aware interview scheduling system that reads real availability across your hiring panel and books slots automatically — no manual back-and-forth
A centralized interview tracker that pulls scheduled interviews from Google Calendar, collects interviewer feedback in one place, and flags missing scorecards before the debrief
An automated candidate communication workflow that drafts confirmation emails, sends pre-interview prep notes, and follows up with next-step messages so you're not writing the same five emails every week
The Starch recipe

Apps, data, and prompts

The combination of Starch apps, the data sources they pull from, and the prompts you use to drive them.

Data sources & config

Starch syncs your Google Calendar data on a schedule so the scheduling app reads live panel availability without manual input. Gmail is connected as a scheduled-sync provider so the Email Triage app drafts and sends candidate-facing messages from your own inbox. Slack is connected as a scheduled-sync provider so debrief reminders fire into the right channel. For any ATS your team uses — Greenhouse, Lever, or Ashby — Starch connects from its integration catalog and queries candidate stage data live when the tracker runs.

Prompts to copy
Build me an interview scheduling assistant that reads the Google Calendars of everyone on the hiring panel for a given role, finds overlapping 45-minute windows in the next 10 business days, and shows me three candidate options with no back-to-back meetings for panelists who requested buffer time
Create an interview loop tracker that shows each open role, the candidates in process, which interview stages are complete, which scorecards are missing, and who I need to chase — update it from Google Calendar events tagged with the role name
Set up an automation that emails interview candidates a confirmation and prep note 24 hours before their scheduled slot, and Slacks me a reminder 48 hours after each debrief if I haven't logged a hiring decision for that candidate
Run these in Starch → or paste them into your favorite agent
Walkthrough

Step-by-step

1 Connect Google Calendar as a scheduled-sync provider in Starch so the agent can read availability for you and every hiring manager on your panel — no manual schedule exports.
2 Connect Gmail as a scheduled-sync provider so Starch can draft and send candidate confirmation emails, prep notes, and next-step messages directly from your account.
3 Connect Slack as a scheduled-sync provider so Starch can post debrief reminders, missing-scorecard alerts, and hiring-decision nudges to your recruiting channel.
4 If your team uses Greenhouse or Lever, connect it from Starch's integration catalog — the agent queries candidate stage and interviewer assignment data live each time the tracker refreshes.
5 Open the Scheduling app from the Starch App Store and configure it with your standard interview formats: 30-min recruiter screen, 45-min technical, 60-min panel debrief. Set buffer rules for panelists who requested them.
6 Tell Starch: 'Build me an interview loop tracker that shows every active candidate by role, their current stage, which interviews are scheduled vs. completed, and which scorecards are overdue — pull scheduled events from Google Calendar and stage data from Greenhouse.' Starch builds the app; you review and adjust the columns.
7 Tell Starch: 'Every weekday morning, check for interviews scheduled in the next 24 hours and send each candidate a confirmation email with prep notes from our standard template. Send it from my Gmail.' Review the first three sends, then let it run.
8 Tell Starch: 'Two days after any calendar event tagged as a debrief, check whether I've logged a hiring decision for that candidate in the tracker. If not, Slack me in #recruiting with the candidate name and role.' This closes the loop on stalled decisions.
9 Use the Task Manager to capture the manual exceptions — the hiring manager who insists on a phone call to set up interviews, the candidate who needs a rescheduled slot. Assign P1/P2 priorities and due dates so they don't fall through while the automation handles the routine cases.
10 Before each weekly recruiting sync, tell Starch: 'Summarize all interview activity from the past seven days: how many screens, panels, and debriefs ran, which roles have candidates pending a decision, and which interviewers have submitted fewer than 50% of their scorecards this quarter.' Use the output to run the meeting in 15 minutes instead of 40.
11 When you're ready to make an offer, tell Starch: 'Draft an offer email for [candidate name] for the [role] position at [salary/equity], referencing the offer letter template in our Notion workspace, and send it from my Gmail.' Starch pulls the Notion template and drafts the email for your review.
12 After each hire closes, tell Starch: 'Log the time-to-fill for [role] — days from first interview to offer accepted — and add it to the recruiting metrics table.' Over time, you have actual data to bring to the CEO instead of anecdote.

See this running on Starch

Connect your tools, describe what you want, and the agent builds it. Closed beta is free.

Try it on Starch →
Worked example

Q2 2026 — Head of Growth Hiring Loop

Sample numbers from a real run
Recruiting screens scheduled14
Panel interviews coordinated across 5 interviewers8
Scorecards missing at debrief (caught by Starch alert)3
Candidate emails drafted and sent automatically22
Hours of manual scheduling saved over 3-week loop6

You opened the Head of Growth role on April 7. By April 9, Starch had the scheduling app live, reading the Google Calendars of your CMO, two Growth team ICs, and the CEO — all connected as scheduled-sync providers. When the recruiter moved a candidate to panel stage in Greenhouse, Starch queried the ATS live, identified three overlapping 45-minute windows in the next eight business days where none of the five panelists had back-to-back conflicts, and surfaced them for your review. You approved, Starch sent the candidate a confirmation email with prep notes from Gmail within the hour. Two days after the April 18 debrief, Starch pinged #recruiting in Slack: 'No hiring decision logged for Sarah Chen / Head of Growth.' Three scorecards had been missing at debrief — Starch had flagged that 48 hours earlier, so you were already chasing the two ICs. The CEO submitted her scorecard before the debrief started. Offer went out April 23. Time-to-fill: 16 days. Your previous comparable hire took 31.

Measurement

How you'll know it's working

Time-to-fill per role (days from first interview scheduled to offer accepted)
Scorecard submission rate — percentage of interviewers who submit within 48 hours of their interview
Scheduling turnaround time — hours between 'move to panel' in ATS and candidate confirmation sent
Interview-to-offer conversion rate by role and interviewer panel
CEO and exec calendar hours spent on interviews per quarter (are they over-indexed on early-stage screens?)
Comparison

What this replaces

The other ways teams handle this today, and how the Starch version compares.

Calendly alone
Handles single-person booking cleanly but has no panel conflict detection, no ATS integration, and no post-interview automation — you still do the coordination manually.
Greenhouse or Lever native scheduling
Good if your entire team lives in the ATS, but scheduling logic is rigid and it doesn't touch Gmail drafts, Slack alerts, or cross-functional calendar conflicts for exec panelists who aren't in the ATS.
Google Sheets + Zapier
Flexible and cheap to start, but every new rule is another Zap to maintain, there's no natural-language authoring, and the tracker falls apart the moment someone renames a calendar event slightly differently.
Notion database + manual process
Great for documentation but entirely passive — Notion doesn't chase down missing scorecards, draft emails, or read calendar availability; you still do all of that by hand.
On Starch RECOMMENDED

One platform — scheduling, task manager, founder inbox all running on connected data. Setup in plain English; numbers stay current via scheduled syncs and live agent queries.

Try it on Starch →
FAQ

Frequently asked questions

Does Starch actually read my hiring managers' calendars, or just mine?
Starch syncs any Google Calendar account that the calendar owner connects. In practice, you'd ask each panelist to connect their Google Calendar through Starch — it takes about two minutes. Once connected, the scheduling app can read their availability directly. If a panelist won't connect, you can enter their availability manually or share their calendar with your account and have Starch read it through your connection.
We use Ashby, not Greenhouse or Lever. Does this work?
Starch connects to 3,000+ apps through its integration catalog, plus any website through browser automation. If Ashby is reachable through the integration catalog, the agent queries it live. If not, Starch can automate Ashby through your browser — no API required. Either way, the tracker can pull candidate stage data; the setup process tells you which path applies.
Can Starch send emails that look like they're from me, not a generic noreply address?
Yes. Starch connects directly to Gmail as a scheduled-sync provider, which means it drafts and sends from your actual Gmail account. The candidate sees your email address, not a platform address. Note: the Gmail OAuth consent screen currently shows the underlying connector's name rather than 'Starch' — that's on our roadmap to fix, but it's a cosmetic issue, not a functional one.
What if an interviewer changes their availability the day before?
Starch syncs Google Calendar on a schedule, so it catches updates regularly. If a panelist blocks time after a candidate slot is already confirmed, Starch won't automatically unbook that slot — it surfaces the conflict and alerts you so you can decide whether to reshuffle or leave it. Fully autonomous rebooking without human review is something you'd need to explicitly configure and sign off on.
Is Starch SOC 2 certified? We have a security-conscious People team.
Not yet — Starch is not SOC 2 Type II certified today. That's the honest answer. If your People team has a hard requirement for SOC 2 before connecting calendar and email data, that's a real constraint to weigh. It's on the roadmap, but we won't tell you it exists before it does.
Can I build a structured feedback form for interviewers inside Starch?
Yes. Tell Starch: 'Build me an interviewer feedback form with fields for candidate name, role, interview type, a 1-5 rating on each of our four competencies, and a hire/no-hire recommendation — and log each submission to the interview tracker.' Starch builds the form and the logging logic. It's not a polished form tool like Typeform, but it's connected to your data, so the scorecard lands in the tracker automatically instead of in someone's email.

Ready to run run an interview loop on Starch?

Request closed-beta access. Everything is free during beta.

You're on the list! We'll be in touch soon.