Back office ops · Production

GenV: agentic workflow for extracting structured insights from Google Meet recordings using Vertex AI Gemini

The problem

Crucial details from video meetings get lost after calls end; manually scrubbing through recordings to find decisions, action items, and context is tedious and inefficient, and generic note-taking apps are insufficient.

Workflow diagram · grounded in source
1
Locate recent recordings
trigger
“It mounts your Google Drive to access the specified path for Meet recordings (e.g., MyDrive/Meet Recordings). It identifies video files modified within a recent period (e.g., last 30 days, configurable via SINCE_DAYS)”
2
Upload to cloud storage
integration
“the script uploads these video files from Drive to a designated Google Cloud Storage (GCS) bucket if they don't already exist there. This prepares the video for analysis”
3
Gemini multimodal analysis
ai_action
“For each video file (referenced by its GCS URI), the script sends a request to the Gemini model... specifying the response_mime_type as application/json and providing the response_schema (our MeetingInsight Pydantic model). This multimod…”
4
Generate reports and save data
output
“It generates readable Markdown summaries, organizing the extracted information under relevant headings (Summary, Q&A, Projects, Action Items, Decisions, etc.). It also saves the raw structured data to a jsonlines file for potential downs…”
Reported outcome

GenV automates structured knowledge extraction from meeting recordings, providing rapid summarization, action item capture, decision tracking, and knowledge retrieval while saving significant time compared to manual review.

Reported metrics
Time saved vs manual reviewsignificant time
Reported stack
GenVGoogle ColabGoogle Cloud StorageVertex AI APIVertex AI's Gemini modelsGoogle MeetGoogle Drive
Source
https://mlops.community/blog/genv-an-agentic-workflow-for-actionable-insights-from-google-meet-recordings
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

GenV automates structured knowledge extraction from meeting recordings, providing rapid summarization, action item capture, decision tracking, and knowledge retrieval while saving significant time compared to manual r…

What tools did this team use?

GenV, Google Colab, Google Cloud Storage, Vertex AI API, Vertex AI's Gemini models, Google Meet, Google Drive.

What results were reported?

Time saved vs manual review: significant time (source-reported, not independently verified).

How is this back office ops AI workflow structured?

Locate recent recordings → Upload to cloud storage → Gemini multimodal analysis → Generate reports and save data.

Related back office ops cases