John Farrall

Add the Alternative Data Weekly to the long list of famous 39s! @hasek_dominik @DaveParker39fdn @ScuderiaFerrari @LarryCsonka @substack listed the ADW #39 on the Rising Bestseller list! #altdata #alternativedata #ai #datacommercialization Sub here: farrall.substack.com

@twtayaan Great visual. Thanks for sharing. Focused on identifying problems earlier in the cycle + making cost more predictable/transparent.

The 3 Pillars of Observability: ▶️ Metrics: → Scrapes data points like CPU usage and response times to measure performance. → Provides a high-level overview of system health and behavior over time. → Uses Prometheus for efficient collection and visualization via dashboards. ▶️ Logs: → Captures detailed records of system activities, events, and error messages. → Offers a historical view to help diagnose and troubleshoot specific incidents. → Uses Loki to aggregate and index log data from multiple sources seamlessly. ▶️ Traces: → Tracks and visualizes the end-to-end flow of requests across microservices. → Captures latency data to identify exact bottlenecks in the request path. → Uses Jaeger to diagnose performance issues in complex, distributed systems. Which pillar are you currently focusing on improving in your stack?

See you in Minneapolis, May 21-22, 2026! events.linuxfoundation.org/observability-… #o11ysummit @linuxfoundation @SymetryML

@grafana’s 2026 Observability survey: “Nearly everyone (95%) says it's important for AI to show its reasoning” The good news: @SymetryML delivers root cause context instantly, dramatically reducing MTTR and cognitive load on engineering teams. Engineers receive the What, Why, and Where of the incident before they've opened a single dashboard. Less investigation time. Faster resolution. #Observability #SRE #o11y

What does @SymteryML NOT do? - Storeraw telemetry. -Replace your existing observability tools. - Require changes to instrumentation, agents, or dashboards.

@SymetryML deploys as a lightweight, read-only sidecar into your existing telemetry pipeline. · no rearchitecture. · no new agents. · no disruption to your current stack. SymetryML connects at the collection layer, receiving a parallel copy of your telemetry stream before it reaches storage or any downstream observability tool. Your existing pipelines, dashboards, and alerting workflows remain completely unchanged.

@grafana's 2026 Observability survey: “Nearly everyone (95%) says it's important for AI to show its reasoning” The good news: @SymetryML delivers root cause context instantly, dramatically reducing MTTR and cognitive load on engineering teams. Engineers receive the What, Why, and Where of the incident before they've opened a single dashboard. Less investigation time. Faster resolution.

@grafana's 2026 Observability Survey: 75% of respondents indicate assistance with Root Cause & Correlation Analysis is Critical or Very Valuable. @SymetryML's Detection Summary Card delivers the What, Why, and Where of every incident. Engineers arrive at the incident fully oriented. No manual correlation. No starting from scratch. Less investigation time. Faster resolution. #Observability #SRE #o11y

@grafana's 2026 Observability Survey: The top three most important criteria for selecting new observability tools: - Cost - Ease of use - Interoperability @SymetryML is a read-only sidecar. No rip & replace. No new agents. No rearchitecture. No raw data stored. Fixed compute cost regardless of data volume. No longer are teams forced to choose between full visibility & sustainable spend. Plug in. Get smarter. Nothing else changes. #Observability #FinOps #SRE

Really enjoyed the Data Exchange podcast episode with @lgavish (CTO & Co-Founder with @BM_DataDowntime of @montecarlodata). One of the clearest takes I’ve heard on how data observability is evolving as AI becomes embedded in real production systems. @bigdata's highly recommended @GradientFlowR blog post that instigated this conversation: Beyond Black Boxes: A Guide to Observability for Agentic AI. Lior made two points that resonated: 1- Observability is going to be about ensuring trust in AI workflows. The opportunity is huge. 2- It is hard to extract insight from telemetry (~23:30). gradientflow.substack.com/p/are-your-ai-…

@grafana's 2026 Observability Survey: Alert fatigue is the #1 obstacle to faster incident response. Alert fatigue is a threshold problem. You can't tune your way out of it. @SymetryML replaces static threshold rules with behavioral drift detection. Alerts fire only because something actually changed. #Observability #DevOps #SRE

@grafana's 2026 Observability Survey: ~92% of teams see real value in AI that catches anomalies before they cause downtime. That's not a feature request. That's a mandate. @SymetryML runs upstream of storage, detecting behavioral drift the moment it emerges. Not after. Not on query. Live. #Observability #SRE #o11y

@grafana's 2026 Observability Survey: Alert Fatigue is the #1 obstacle to faster incident response time. @SymetryML replaces static threshold rules with behavioral drift detection. Alerts fire only because something actually changed. Less investigation time. Faster resolution. #Observability #FinOps #SRE

The hardest part of on-call isn't the 2am pages. It's the 2am pages that turn out to be nothing. Again. Engineers don't burn out from hard problems. They burn out from false positive noise. Behavioral intelligence over static thresholds. That's the direction. @SymetryML #SRE

Nobody starts a business being excited about spending $500k/year on observability tooling. It happens incrementally. One integration. One new service. One cardinality spike that never came back down. By the time you notice, it's a line item your CFO is asking about. There's a better architecture for this. #o11y

Look forward to seeing everyone at the Observability Summit, May 21-22. register.linuxfoundation.org/WO3YYy?sms=7&c…

The incident postmortem is always the same story: "The signals were there. We just didn't see them in time." Static thresholds can't see behavioral drift. They only see the moment something breaks, not the hours & days of subtle change that led there. That's the gap @SymetryML is built to close. #SRE

High cardinality isn't a bug in your system. It's a feature of your business growing. The problem is that it drives costs higher. Your bill scales exponentially with your linear success. @SymetryML handles high-cardinality telemetry. Business growth does not mean higher costs. #o11y

Be honest: how much of your on-call rotation is triaging alerts that turn out to be nothing? 10%? 40%? More? That's engineering time that should be on features. The static threshold model wasn't designed for the complexity of modern distributed systems. Curious what people are actually experiencing. Drop a number below. 👇

Alternative Data Weekly #282 has been published! Theme: We’re Entering the Era of Programmatic Data Consumption farrall.substack.com/p/alternative-…