Nova Copilot

Engineering insights and product updates

Field reports on SRE, agentic AI, observability, security, and building reliable systems at scale. Written by practitioners who spent years on-call at hyperscalers, then built the platform they wished they had.

The Nova AI Ops blog covers the hard problems of modern SRE in 2026, reducing alert fatigue without missing real incidents, cutting MTTR from hours to minutes with agentic AI, deploying OpenTelemetry-native observability at scale, hardening the software supply chain with SBOMs and SLSA, and writing runbooks AI agents can actually execute. Every article is practical, opinionated, and grounded in real incidents we or our customers have lived through.

Popular topics

By year
Browse by topic RSS 19 categories · 100 articles

Featured

April 2026

Agentic SRE: The Operating System for Autonomous Site Reliability

The complete category guide. Detect, diagnose, decide, remediate, audit, learn. How 100 specialized AI agents across 12 teams redefine site reliability.
Apr 24, 202614 min readRead article →

Editor’s Picks

Two must-reads

Agentic SRE vs AIOps: The Architectural Differences That Matter

Every AIOps vendor is about to ship an agentic marketing page. The clean architectural test plus a side-by-side incident walkthrough.
Oct 3, 202610 min readRead →

AIOps Platforms Buyer’s Guide 2026

Eight must-have capabilities, seven red flags, and the top six vendors. Pricing, ROI calculators, and deployment checklists for 2026.
Sep 24, 202615 min readRead →

Latest Articles

100 articles

Kubernetes Ingress Controllers Compared 2026

nginx, Traefik, HAProxy, Envoy/Contour, AWS ALB Controller. The four-criteria comparison and the migration cost.
Oct 11, 20269 minRead →

On-Call After-Hours Policy: Boundaries That Stick

Boundaries during off-hours protect the on-call from drift into ‘always on.’ The four boundary patterns and the policy that makes them durable.
Oct 30, 20269 minRead →

Honeycomb vs Datadog: Observability Approaches Compared

Honeycomb is event-stream-first; Datadog is dashboard-first. The two approaches are complements as often as competitors.
Nov 25, 20269 minRead →

Heroku vs Vercel vs Render: Modern PaaS Compared

Heroku set the standard; Vercel and Render took different lessons from it. The four-criteria split that picks correctly, and the migration cost from Heroku in 2026.
Nov 27, 20268 minRead →

Falco vs Tetragon: Runtime Security Tools Compared

Falco uses syscall hooks; Tetragon uses eBPF. The performance, observability, and policy-enforcement tradeoffs that decide which is right for your environment.
Dec 1, 202610 minRead →

Crossplane vs Terraform: Infrastructure-as-Code in 2026

Terraform is the legacy default; Crossplane is the Kubernetes-native challenger. The four scenarios where each wins, and the hybrid pattern that uses both well.
Dec 1, 202611 minRead →

Error Budget Burn-Rate Alerts: The Math Behind Modern SLOs

From percentages to multi-window burn rates: why fast-burn and slow-burn alerts beat threshold rules, the specific equations, and a copy-paste Prometheus example.
Dec 19, 20269 minRead →

Distributed Tracing Sampling Strategies That Don't Lie

The four sampling strategies (head-based, tail-based, adaptive, error-priority), what each one biases for and against, and the hybrid most production teams converge on.
Sep 13, 20266 minRead →

Customer-Facing Incident Comms Templates

The three core templates (acknowledged, in-progress, resolved), what to put in each slot, the tone rules, and the four words to never use in incident comms.
Sep 11, 20265 minRead →

Automation Debt: The Slow Drag You Cannot See

How automation debt accrues, the tracking spreadsheet that surfaces it, the four classes of debt (one-off scripts, undocumented procedures, vendor-locked tooling, manual-only paths), and the order to pay them down.
Aug 12, 20265 minRead →

Graceful Degradation: How a Site Stays Half-Up

The four degradation patterns (read-only mode, cached fallback, degraded UI, drop-the-feature), what each costs to build, and the order to add them as your service matures.
Aug 9, 20266 minRead →

What Is an Agentic SRE Agent? A Technical Breakdown

The five components every production-grade SRE agent needs: identity, memory, tools, policy envelope, trust score.
Oct 3, 202610 minRead →

Datadog Alternatives 2026: The Complete Comparison

The top 10 monitoring and observability platforms in 2026 compared on capability, pricing, and fit.
Sep 25, 202612 minRead →

PagerDuty Alternatives for Incident Management in 2026

The best PagerDuty alternatives compared: Nova AI Ops, OpsGenie, FireHydrant, Rootly, Incident.io, xMatters.
Sep 25, 202611 minRead →

Best SRE Tools 2026: The Complete Guide

The definitive SRE tooling guide. Monitoring, incident management, automation, on-call, and runbooks.
Sep 14, 202616 minRead →

SRE Best Practices 2026: The Complete Handbook

SLOs, error budgets, toil reduction, on-call management, post-mortems, automation, observability.
Sep 14, 202618 minRead →

How to Reduce MTTR: A Practical Guide for SRE Teams

Seven proven strategies that take Mean Time to Resolution from hours to minutes, with real AI-driven data.
Sep 11, 202612 minRead →

Alert Fatigue: What It Is and How to Fix It

Alert fatigue is the leading cause of missed incidents. Five proven solutions including AI correlation.
Sep 12, 20269 minRead →

Eliminate Alert Noise: The 2026 Playbook

AI-driven alert correlation reduces volume by 90%+ without missing real incidents. The concrete tactics.
Sep 12, 202610 minRead →

Kubernetes Incident Management 2026

Common K8s failure modes, debugging workflows, auto-remediation patterns, and how AI agents transform K8s SRE.
Sep 12, 202613 minRead →

SLI vs SLO vs SLA: The Three-Letter Acronyms That Actually Matter

The exact distinction between SLIs, SLOs and SLAs, why confusing them costs real money, and a one-page template that disambiguates the three for any service.
Jul 9, 20267 minRead →

Terraform vs Pulumi vs CloudFormation: A Pragmatic 2025 Comparison

Language surface, state model, blast radius, and ecosystem, the four axes that separate these three IaC tools when your infra grows past a proof of concept.
May 4, 202610 minRead →

Prometheus vs InfluxDB vs Grafana Cloud: A Practical 2025 Comparison

A side-by-side on storage model, query language, cardinality ceiling, cost shape, and operational overhead. Plus the two questions that decide which one you actually need.
Apr 13, 202611 minRead →

Vector Search at Scale: Beyond pgvector

What breaks at scale, the index types you need (HNSW, IVF, ScaNN, DiskANN), the sharding patterns, and how to choose between Pinecone, Milvus, Qdrant, and self-hosted options at 100M+ vectors.
Jun 7, 20269 minRead →

Streaming LLM Responses: UX + Latency Math

Why streaming changes user perception, the four latency metrics that matter, the SSE/WebSocket implementation choice, and the failure modes to plan for.
Jul 11, 20266 minRead →

Agentic Reasoning: Tree of Thoughts, ReAct, and Reflection

Tree of Thoughts (parallel branches), ReAct (interleaved reasoning + acting), and Reflexion (self-critique). What each adds, when each helps, and how they combine.
Aug 9, 20267 minRead →

Edge ML: Quantization, Pruning, Distillation

How quantisation, pruning, and distillation compare for edge deployment, the typical accuracy cost of each, and the production stacks (Core ML, TensorRT, ONNX Runtime, llama.cpp).
Sep 11, 20266 minRead →

AI for Scientific Discovery

Where AI has produced verifiable scientific results (protein folding, materials, math), the architecture patterns (search + neural nets), and the limits.
Dec 25, 20265 minRead →

Robotics Foundation Models

What VLA models do, how they unify perception/planning/action, the data scarcity challenge, and the realistic 2026 capability picture.
Dec 24, 20265 minRead →

Datadog vs Dynatrace vs New Relic 2026

Three observability incumbents, three pricing models, and one practical scoring rubric that gets you out of analyst-report purgatory in an afternoon.
Sep 28, 202611 minRead →

Prometheus vs InfluxDB vs VictoriaMetrics 2026

VictoriaMetrics keeps showing up in Prometheus shortlists. Where it actually wins, where Prom still wins, and where neither is the answer.
Sep 23, 202610 minRead →

PagerDuty vs OpsGenie vs Incident.io 2026

Routing, scheduling, lifecycle, post-mortem flow, scored side-by-side on the workflows on-call engineers actually run, not the marketing matrix.
Sep 19, 202610 minRead →

Tracing Tools: Jaeger vs Tempo vs Honeycomb 2026

Three popular distributed-tracing backends, three very different operational profiles. The cost-vs-cardinality tradeoffs and the tier each one wins.
Aug 27, 20269 minRead →

Alert Grouping and Deduplication, Done Right

The four grouping dimensions (service, time window, label, root cause) and how to combine them so 200 raw alerts become a single actionable incident.
Sep 29, 20269 minRead →

Alert Routing: Severity to Owner, Without the Hops

Every minute spent rerouting an alert is a minute on the SLO. The label-driven routing pattern that holds up across mergers, reorgs, and team renames.
Sep 27, 20269 minRead →

Designing Alert Severity Levels

Sev-1 through Sev-4 sounds simple until two engineers disagree at 3am. The single-page rubric that gets every team using the same words.
Sep 22, 20268 minRead →

Actionable vs Informational Alerts

If a human can’t act on it in five minutes, it shouldn’t page. The two-question test that cuts most alert volume by 60% on the first pass.
Aug 28, 20266 minRead →

Kubernetes Ingress Controllers Compared 2026

NGINX, Traefik, HAProxy, Envoy-based, plus the Gateway API question. A scoring rubric so the choice survives the next three K8s versions.
Sep 21, 202610 minRead →

Kubernetes Cost Optimization Playbook

Right-sizing, spot, bin-packing, idle-pod hunting, namespace quotas. Five levers in the order that gets the biggest savings without breaking reliability.
Sep 17, 202611 minRead →

Best Kubernetes Observability Tools 2026

The five tools every cluster needs, the three that overlap, and the AI-native pattern that finally makes pod-level tracing affordable.
Sep 15, 202610 minRead →

Kubernetes GitOps: Argo CD vs Flux 2026

Two GitOps controllers, two different mental models. Repo layout, drift detection, and the multi-cluster patterns each one is actually built for.
Aug 24, 202610 minRead →

Best AIOps Platforms 2026

Twelve AIOps platforms scored on detection, correlation, automation, post-mortems, and total cost of ownership. The clear leaders, and the laggards.
Sep 29, 202613 minRead →

Agentic SRE vs AIOps

A category buyer’s guide. What separates agentic SRE from classic AIOps, and the seven capability lines that decide which one your team needs.
Oct 3, 202611 minRead →

AIOps RFP Template 2026

A vendor-neutral RFP you can paste into a Google Doc. 60 questions, 8 categories, scoring rubric included.
Sep 23, 202610 minRead →

AIOps Pricing Models Explained

Per-host, per-user, per-event, per-GB, plus the “contact us” trap. The five common models and which one usually wins on a 3-year TCO.
Sep 18, 20269 minRead →

AIOps Implementation Timelines

Day-1, week-1, month-1, quarter-1: what every realistic AIOps rollout looks like, and the one milestone that predicts whether the platform will stick.
Sep 17, 20269 minRead →

How to Evaluate AI SRE Vendors

Five live demos that separate real autonomy from rebadged dashboards, plus the reference-call questions that get past marketing.
Sep 15, 202610 minRead →

AIOps ROI Calculation Guide

A real ROI model that the CFO will sign. Tool consolidation, MTTR delta, on-call comp, and the human-time savings nobody wants to put on a slide.
Sep 13, 202610 minRead →

AIOps Vendor Selection Rubric

Twelve weighted dimensions, four-point scoring, single-page summary. Drop names in, get a ranked short list and a defensible decision memo.
Sep 10, 20269 minRead →

Monitoring Platform RFP 2026

The vendor-neutral RFP for observability platforms. 50 questions, scoring rubric, and the “leave-blank” cells that catch over-promised features.
Sep 8, 202610 minRead →

Incident Management Buyer’s Guide 2026

PagerDuty, Incident.io, FireHydrant, Rootly, OpsGenie, plus the AI-native challengers. Scoring rubric for routing, lifecycle, and post-mortem flow.
Sep 7, 202611 minRead →

Observability Platform Buyer’s Guide 2026

Datadog, Grafana, New Relic, Splunk, Honeycomb, plus the open-source stack. Side-by-side scoring on cost, depth, and openness.
Sep 5, 202612 minRead →

AIOps Migration Guide

Datadog out, Nova in, or whichever direction you’re going. The dual-run pattern, the data-portability checklist, and the cutover script.
Sep 3, 202611 minRead →

AIOps: Build vs Buy in 2026

The four costs you forget when you build, the three you don’t see when you buy, and the small set of orgs where building still makes sense.
Aug 31, 202610 minRead →

Real Outage: A Database Failover That Failed Over

A 24-hour data-locality incident sparked by a planned failover that took longer than the timeout. The split-brain risk, and the runbook redesign.
Sep 13, 202611 minRead →

Real Outage: Kafka Consumer Rebalance Storm

A rolling restart on a 240-consumer group triggered 9 minutes of continuous rebalancing. The session-timeout vs heartbeat math that fixed it.
Aug 25, 202610 minRead →

Real Outage: A Redis Cluster Split-Brain

A 90-second network blip created two primaries. The 4 minutes of dual writes, the reconciliation script, and the quorum config that came out of it.
Aug 23, 202610 minRead →

Single-Shot vs Iterative Agents for Incident Response

Some incidents need one model call with the right context. Some need iterative reasoning over many turns. The cost and latency math that picks the right shape per incident type.
Jul 28, 20265 minRead →

The Agent Cost Bomb: Pre-emptive Token Budgets

One stuck agent can burn $400 in an hour. The budget enforcement layer that stops it before it does, plus the alerting that wakes you up if budgets blow up across runs.
Jun 30, 20265 minRead →

The Action-Limit Pattern: Capping What an Agent Can Do

Hard caps per run, per service, per minute. The cap dimensions that matter, sensible defaults, and the dashboard that catches caps quietly hitting in production.
Jun 27, 20265 minRead →

The Action-Stagger Pattern: Throttling Agent Side Effects

Bunched actions amplify blast radius. Stagger them and you get observability between each. The throttle policy, with code, that turns a thundering herd into a measured walk.
Jun 19, 20265 minRead →

Distributed Tracing for Multi-Agent Systems

When five agents collaborate, a single trace is the only way to debug. The instrumentation, the span layout, and the queries that find the slow specialist.
Jun 14, 20265 minRead →

The Agent Run Timeline: Building a Replay UI

A timeline you can scrub. The web component, the data model, and the keyboard shortcuts that turn an opaque run into something a junior SRE can debug.
Jun 12, 20265 minRead →

The Agent Audit Log: What Goes In, What Comes Out

Auditors will ask. The audit-log schema that satisfies SOC2, PCI, and your own future investigation, with retention policy and access-control notes.
Jun 4, 20265 minRead →

Tracking Tool-Call Failures: A Dashboard That Matters

Tool failures cause more agent regressions than model regressions. The five panels, the alert thresholds, and the runbook entry that brings the on-call up to speed.
Jun 2, 20265 minRead →

Multi-Agent Workflows for Postmortem Generation

One agent gathers data. One writes. One reviews. One files. The workflow, with the inter-agent messages typed and bounded.
Apr 5, 20265 minRead →

Prometheus vs VictoriaMetrics: 2026 Decision

Prometheus is the standard; VictoriaMetrics is the high-performance alternative. The decision criteria with concrete numbers.
Jul 18, 20264 minRead →

The PromQL Patterns Checklist Every SRE Should Know

Twelve PromQL patterns that cover 80% of production queries. The checklist with examples and what each catches.
Jul 14, 20264 minRead →

Loki vs Elastic: 2026 Decision Guide

Loki is cheap and label-driven; Elastic is full-text and powerful. The decision criteria for picking a logging backend in 2026.
Jun 7, 20264 minRead →

The Multi-Window Multi-Burn-Rate Alert

The Google SRE pattern: alert on burn rate over multiple windows simultaneously. Why it works, with the configuration.
Apr 19, 20264 minRead →

SSM vs SSH: 2026 Default for Server Access

SSH still works but is harder to audit. SSM Session Manager replaces SSH for most use cases.
Apr 6, 20264 minRead →

Cloudflare Workers vs Lambda@Edge

Two edge compute platforms. The decision criteria for picking one.
Mar 11, 20264 minRead →

Cloud Provider Egress Fees 2026

Egress fees are gradually decreasing. The 2026 picture and the strategies for cost control.
Feb 11, 20264 minRead →

Grafana Faro vs Other RUM

Faro is Grafana's RUM tool. The decision criteria.
May 23, 20264 minRead →

AWS SAML CLI Tools (saml2aws, aws-sso)

SAML auth for AWS CLI.
Mar 25, 20264 minRead →

zsh vs bash for SREs

Shell choice. Productivity differences.
Mar 17, 20264 minRead →

htop vs btop for System Monitoring

htop is classic; btop is modern.
Mar 1, 20264 minRead →

mosh vs ssh for Unstable Connections

mosh handles network changes.
Feb 16, 20264 minRead →

mitmproxy for API Debugging

mitmproxy intercepts API traffic.
Feb 4, 20264 minRead →

sops for Encrypted Secrets in Git

sops encrypts files for git storage.
Jan 29, 20264 minRead →

Building CLI Tools: Go vs Rust

Choose CLI tool language.
Jan 18, 20264 minRead →

Okta vs OneLogin CLI Auth

SSO tools for CLI access.
Dec 25, 20254 minRead →

tracee and Falco for Runtime Security

Two runtime security tools.
Dec 13, 20254 minRead →

Renovate vs Dependabot

Two dependency update bots.
Nov 18, 20254 minRead →

Error Tracking Tool Decision

Three error trackers compared.
Nov 8, 20254 minRead →

py-spy for Python Performance

py-spy is a sampling profiler for Python.
Oct 19, 20254 minRead →

Prometheus Alertmanager Routing

Alertmanager's tree-based routing. The patterns that work.
Apr 7, 20264 minRead →

Alert Deduplication Strategy

Same incident, multiple alerts. Dedupe early.
Jan 28, 20264 minRead →

Alert Vendor Comparison 2026

PagerDuty, Opsgenie, VictorOps, others. The differences.
Dec 17, 20254 minRead →

Data Retention Policy

How long to keep data. The policy.
Jan 9, 20264 minRead →

Cyber Insurance Engineering

Cyber insurance requires controls. The engineering.
Aug 26, 20254 minRead →

Progressive Delivery Tools

Argo Rollouts, Flagger. Beyond Deployment.
Jan 8, 20264 minRead →

Canary vs Feature Flag

Two ways to reduce deploy risk.
Sep 12, 20254 minRead →

Acknowledgment Time SLA

< 5 min for sev 1.
Sep 8, 20254 minRead →

Connection Multiplexing

HTTP/2 advantage.
Aug 5, 20254 minRead →

Feature: Datadog Integration

Metrics flow.
Apr 27, 20254 minRead →

Careers Update

We're hiring.
Jan 19, 20254 minRead →

AI Pricing Models 2026

Per-token, per-call.
Jan 3, 20254 minRead →

No matches yet

We don't have anything in this slice of the catalog. Try a different topic, year, or clear all filters to start over.

Stay in the loop

Get engineering insights and product updates delivered to your inbox.