Interview feedback collection and reminders
After each interview, an agent tracks which panel members still owe written feedback, sends polite reminders, and assembles submitted notes into one view for the hiring debrief without suggesting any decision.
The use case
Hiring decisions often stall because interview feedback is late, scattered across chat messages, or written days later from memory. When notes are missing, debriefs drift toward impressions rather than the criteria the team agreed on. This agent watches the interview schedule and the applicant tracking system, notices which interviewers have not submitted their scorecard, and sends reminders in Slack at set intervals. When feedback comes in, it collects the notes for each candidate into one view organized by the role's interview criteria, quoting each interviewer rather than rewriting their opinions. It does not combine feedback into an overall verdict or suggest a decision. The hiring manager reads the full set of notes, runs the debrief, and records the decision and reasoning in the applicant tracking system.
Example workflow
- After an interview ends, the agent checks Greenhouse to see whether each panel member has submitted a scorecard.
- If feedback is missing after a set number of hours, it sends the interviewer a short Slack reminder with a link to the scorecard.
- It repeats the reminder once more and then notifies the recruiter about any feedback still outstanding.
- When all feedback is in, it assembles the notes by interview criterion, quoting each interviewer and marking which criteria nobody covered.
- The hiring manager reviews the full notes, holds the debrief, and makes the decision with the panel.
- The decision and its reasons are recorded by the hiring manager in the applicant tracking system, and the agent logs only reminder dates.
Tools
- Greenhouse
- Slack
- Google Calendar
- Claude
- Zapier or Make
Risks and guardrails
- An assembled view could tilt the outcome if it summarizes opinions unevenly, which matters under employment-discrimination law — guardrail: the agent quotes interviewers directly, never produces a verdict, and every hiring decision goes through human review by the hiring manager and panel.
- Interviewers may copy the agent's structure instead of writing their own view — guardrail: the agent only collects feedback after it is submitted and never pre-fills scorecards.
- Reminders could feel like pressure and lead to rushed notes — guardrail: reminder timing and count are set by HR and kept to a small number.
- Feedback could be shared beyond the hiring team — guardrail: the assembled view stays in the applicant tracking system with the same access rules as the scorecards.
Related recipe
Weekly KPI Digest from Google Sheets with Source-Cited Numbers: An agent reads your defined metrics from Google Sheets, compares them to last week and targets, and drafts a plain-English digest where every number cites its cell. The owner approves before it goes to Slack or email.
Want it set up for you? Request a quote.
Sources
Written by Agents Lookup. Credits and official pages for the tools this use case names:
- Greenhouse (official page): support.greenhouse.io/hc/en-us/articles/52090956642075-Greenhouse-MCP (checked 2026-11-19)
- Slack MCP server (official page): docs.slack.dev/ai/slack-mcp-server (checked 2026-11-09)
- Google Calendar (official page): developers.google.com/workspace/calendar (checked 2026-11-19)
- Claude (official page): claude.ai (checked 2026-11-09)
- Zapier MCP (official page): zapier.com/mcp (checked 2026-11-09)
- Make MCP server (official page): developers.make.com/mcp-server (checked 2026-11-09)