Harsh Yadav.

Contractors kept opening tickets regarding payment delays. Every payment status has a specific meaning and timeline, but nowhere in the product did we explain them in one place.

I wrote an article to capture different statuses, timelines, and what to do at each stage. Then I worked with engineering to show it as a card on the Contractor payments page so contractors saw it before opening a ticket. Ticket volume on payment-status questions dropped.

Sharing this because not every customer problem needs a product fix. Sometimes it’s a knowledge gap, and providing the right information is the fix.

Two automations I built for the ops team.

• Live DB → Google Sheets sync (Borderless). The team needed accurate data for reporting, reconciliations, and other operations. Without live data, they were working from stale exports — manual updates, occasional errors. I built an N8N flow that synced data from the database to a Google Sheet on a schedule. That data became the foundation for further automations and tools that reduced operational error and inefficiency.
• Slack notifications for compliance events (GEOS). An enterprise client required compliance-event notifications in Slack. I built the workflow that listens for the right events and routes them to the right channels in their workspace.

Anything that affects customers — directly or indirectly — is a problem worth solving.

Before we integrated an e-signature workflow, I used AI to go through the full API docs and identify which endpoints we’d actually need for our use case. Then I built internal reference documentation that included the endpoints we use, verified examples, and error responses. I also used AI + cURL to validate the responses against the live API.

I know Postman does similar work and I believe it’s in your stack. We didn’t have it at GEOS at the time, so this was an alternative — and an example of how AI can make a PM more useful to the engineering team and help them ship faster.