DBTB INT Alan Ngai r
summary
Claims that poor ops telemetry changes data and model outputs, and that assembling end-to-end monitoring across pipelines fixes it by letting teams organize your data by concern and cross-correlate metrics in one place. Recommends using hosted tools (OpsClarity style) rather than building everything, notes a small team of five engineers can wire up cloud data, and points to reproducing Netflix-style evaluation as a concrete exercise—practical builds that validate when anomalies drop and predictions improve.
AI-written summary · about
related talks
12mHarnessing Chaos Engineering to Build Robust Systems
31mLeveraging Data Output for Architectural Optimization
01mObservability
34mObservability 2.0: Shifting Left From Reactive Mon...
26mDriving O11y x to Nx: Probing Bullet Holes and not Breaking the Bank
30mPanel: Telemetry That Matters
37mBetter Reliability Through Observability and Experimentation
47mWithout Data, you're just another Person with Opinions