The Handoff studio forced us to write the “last known good” note our NOC had been improvising for months. Dense, but worth the evening sessions.
Flagship course
Metric Canvas Studio
Five weeks of studio practice for analysts and ops leads who need a Data Visualization and Analytics Platform for Streaming Metrics Intelligence they can actually hand over at 2 a.m.
Learning outcomes
- Map a stream’s decisions before choosing chart types.
- Build a canvas hierarchy that separates pulse lanes from portraits.
- Set thresholds that respect seasonal baselines instead of fixed panic lines.
- Leave annotation packs that make night-shift handoffs boring — in a good way.
- Critique boards for vanity aggregates and silent lag.
Modules
- Stream inventory. Sources, lag, ownership, and the questions each metric must answer.
- Canvas grammar. Placement, density, and when sparklines lie.
- Latency Windows. Freshness labels, window choice, and replay critique.
- Threshold memory. Rolling baselines, quiet hours, and alert debt.
- Handoff studio. Annotation languages and a final production board review.
Instructor
Amira Kho — former regional observability lead who rebuilt incident boards for logistics and SaaS teams across Malaysia and Singapore. Amira facilitates Cascade critiques and Observatory intensives.
Informational pricing
Cascade cohort seat: RM 2,450. Observatory team package listed on the pricing page. No checkout on this site.
FAQ
Do I need a specific BI or observability vendor?
No. Exercises use vendor-neutral canvases. You bring screenshots or exports from your stack; we focus on judgment and layout.
How much time should I reserve weekly?
Plan for about four hours: one live session plus studio homework. Teams sharing a board often need an extra hour of sync.
What is a real limitation of this course?
We do not provision streaming infrastructure or write pipeline code. If your events are not yet queryable, the studio cannot invent a warehouse for you — you will need platform engineering support in parallel.
Is there a refund window?
Yes. See the refund policy for eligibility and timelines.
Threshold memory was the module that stuck. I would have liked more multi-tenant examples — our Cascade cohort skewed single-product — yet the critique still cleaned two noisy tiles.