Lessons from an Asset Manager’s First Embedded SRE
summary
A 12‑month embedded SRE trial inside a trading team selects six critical services, uses Grafana and queue‑depth metrics, catalogs alerts in a wiki table, cuts noisy notifications from thousands/week to ~50, holds the pager short‑term to rebuild trust, runs about 100 retrospectives with 3,000 views and uses GenAI to draft timelines, rejects error budgets because finance has zero tolerance for delays, and achieves a sustained 30% annual incident reduction over three years.
AI-written summary · about
chapters (6)
- 0:00Embedding SRE into teams
- 3:27Engagement contract
- 7:18Trust via alerting cleanup
- 15:51Incident retrospectives
- 22:14Supportability in design
- 27:56Takeaways for embedded SREs
related talks
43mFrom "Keeping the Lights On" to "Designing the LEDs": A Detailed Review...
23mSite Reliability Evangelism: Practice Start-up within an Established...
24mHow to Destroy an SRE
27mMean Time To SRE: A DevOps story from adidas
21mMove to Integrated SRE Ops at a Retail organization
52mPersistent SRE Antipatterns: Pitfalls On the...
47mPersistent SRE Antipatterns: Pitfalls on the Road to Creating a Successful
26mTechnology and Cultural Transformation using Site Reliability Engineering