Instatus
Instatus is a status page product serving over 10 million visits per month. Customers include Sketch, T2, Harvard University, Siemens, and Yum! Brands. I joined as a full-stack engineer working across product design, frontend, backend, database, and AWS infrastructure, and I own the chat integrations: Slack shipped, Microsoft Teams in progress.
- Monthly visits, product-wide
- 10M+
- Chat integrations owned
- 2
- Role
- Full-stack engineer
- Period
- 2024 – Present
- Stack
- Node.js, Postgres, Prisma, GraphQL, Redis, AWS, AWS S3, Docker, Sentry, TypeScript, Next.js, React, Figma
- Site
- instatus.com
Context#
Instatus is the status page product engineering teams use to communicate uptime and incidents to their customers. It serves over 10 million visits per month, with customers including Sketch, T2, Harvard University, Siemens, and Yum! Brands. I work across the full stack, with scope that touches product design, frontend, backend, database, and AWS cloud infrastructure.
Owning the Slack integration#
I oversaw the Instatus for Slack integration end to end, from Figma design through to shipped implementation. The rebuild added AI features for incident workflows and shipped a meaningful set of UX improvements on top of a redesigned surface.
Owning the full slice meant making product calls on the design side, carrying them through to the integration code, and then making sure they held up under real customer use. Chat is where incident response actually happens, so the bar is that a responder never has to leave the channel to run the incident.
Microsoft Teams#
I am currently overseeing the Microsoft Teams integration, bringing the same incident workflows to the platform where a large share of enterprise customers already work.
Teams is not a reskin of Slack. The app model, message surfaces, and permission story are different enough that the interesting work is deciding what to keep identical for consistency and what to rebuild to feel native. The Slack rebuild gave us a clear picture of which parts of the workflow actually carry weight, which is what the Teams version is built around.
Reliability: uptime and incidents#
I reworked core systems including uptime percentage calculations and incident handling. Customer trust stakes a lot on these numbers being right, so the goal was correctness, predictable behavior under load, and clear handling of edge cases in time-series aggregation.
On-call, monitoring, and templating#
Shipped improvements across on-call scheduling, monitoring, incident channel creation, and templating. The work spans GraphQL APIs, Redis-backed caching layers, and AWS S3 for media and asset storage.
Testing and reliability patterns#
Established Prisma-based unit, integration, and end-to-end testing patterns across the codebase, and contributed to backend reliability and performance optimization. The throughline is making changes safe to ship in a product where every visible inconsistency erodes trust.