Pumoxi

This meeting summarizer automatically summarizes meeting transcripts into structured, executive style summaries.

---

How it works

1. Input: Paste raw meeting transcript (from Zoom, Teams, notes, etc.).
2. Process: The summarizer extracts key points, actions, and next steps.
3. Output: A polished summary in bullet-point format for easy sharing or documentation.

Input Requirements

• Format: Plain text
• Max length: 1500 characters
• Language: English
• Sensitive data: We don’t store any data, but please avoid any PII / sensitive data.

Sample Input

Below is a sample input generated with Generative AI, all names mentioned are fictional.

Meeting Title: Q2 Platform Stability Review
Date: 2025-06-22
Attendees: Kay, Martin, Jess, Anand, Emily, DevOps team reps, PMs

Jess late 10 mins, joined at 10:12
platform incidents recap — 3 major (April outage, May slow API, June DB failover issue)
Kay: root cause of June issue was misconfigured failover — need to automate validation
Martin mentioned we still don’t have alerting on read replica lag?? (follow up!)
Anand: cost impact of May incident, extra $11k infra cost due to autoscaling thrash
someone (Emily?) raised backlog of tech debt — escalated from devs
Kay suggested setting a baseline SLO for DB response time (200ms p95?)
Emily says BI team’s weekly job keeps timing out — CPU contention on shared nodes
Jess: talk to infra team about splitting job cluster vs app workloads
Martin: reminder to close loop w/ support team, still waiting on RCA doc for April outage
TODOs:
Anand to draft DB failover test runbook by Friday
DevOps to check alerting gaps + read replica metrics by EOD Wed
Jess to ping infra leads re: BI job separation
Kay to propose SLO draft next week

Sample Output

Below is a sample output based on above input, all names mentioned are fictional

Summary:
Meeting Summary: Q2 Platform Stability Review

Date: June 22, 2025
Attendees: Kay, Martin, Jess, Anand, Emily, DevOps team reps, PMs

Key Points:


1. Platform Incidents Recap:
– April Outage: Pending root cause analysis (RCA) document.
– May Slow API: Cost impact of $11k due to autoscaling thrash.
– June DB Failover Issue: Root cause identified as misconfigured failover; need for automated validation.


2. Action Items:
– Martin: Follow up on the lack of alerting on read replica lag.
– Anand: Draft a DB failover test runbook by Friday.
– DevOps: Check alerting gaps and read replica metrics by EOD Wednesday.
– Jess: Discuss with infra team about splitting job clusters vs. app workloads and send a calendar placeholder for the next review.
– Kay: Propose a draft for a baseline SLO for DB response time (200ms p95) next week.


3. Technical Debt:
– Emily highlighted a backlog of technical debt escalated from developers.


4. BI Team Issues:
– BI team’s weekly job timing out due to CPU contention on shared nodes.

Next Steps:
– Next Review: Mid-July (date to be determined).

Next Meeting:
– Jess to send a calendar placeholder for the next review meeting.

Reference

• Meeting summarization and action item extraction with Amazon Nova