October 2026 Summaries
33 posts from New Relic
Filter
Month:
Year:
Post Summaries
Back to Blog
Turo has expanded from local AI coding assistants to cloud-based AI agents integrated into production workflows to address operational challenges in its microservice and Kubernetes environment. As AI-generated code accelerated development, the company found that code review, deployment pipelines, and production incidents became major bottlenecks, reflecting broader industry concerns about increased incidents and senior-engineer firefighting after AI-assisted releases. By centralizing infrastructure metrics, application logs, Kubernetes data, deployment events, and feature-flag information in New Relic, Turo gives AI agents live operational context through the Model Context Protocol server. During a critical overnight incident, an agent reportedly queried telemetry, traced service and log relationships, identified the root cause, and produced an actionable diagnostic report within 60 seconds, before an on-call engineer logged in. Turo’s approach emphasizes that production observability and deployment context are important for using AI agents safely in incident response, release validation, and system design.
Oct 29, 2026
1,038 words in the original blog post.
New Relic has introduced NRQL Macros, reusable named shortcuts for NRQL expressions that aim to reduce query complexity, establish a single source of truth, and simplify bulk updates. Available through the Administration section or API, macros are created and edited centrally, then referenced in queries with a dollar sign followed by the macro name. They can encapsulate lengthy query components such as WHERE or WITH clauses, standardize differently named attributes that represent the same data, and apply shared changes across dashboards with many widgets, such as excluding test users. By centralizing recurring logic, the feature is intended to make NRQL queries more readable, maintainable, and efficient to update.
Oct 28, 2026
592 words in the original blog post.
New Relic announced at GitHub Universe 2026 new integrations for GitHub Copilot and the GitHub Model Context Protocol Server intended to improve reliability as AI-assisted coding accelerates software delivery. Citing its 2026 State of AI Coding report, the company says that although 94% of engineering leaders view AI-generated code as higher quality during review, 78% have experienced more production incidents after deployment. New Relic Change Tracking for GitHub Copilot evaluates pull requests using production telemetry, posts monitoring plans, assesses deployments at six intervals from 10 minutes to 72 hours, and can create GitHub Issues and Copilot-generated fix pull requests when regressions occur. New Relic Autopilot for GitHub MCP Server combines repository history, releases, pull requests, and commit data with live operational signals to help identify the code changes associated with incidents and reduce troubleshooting time. New Relic Preflight for GitHub Copilot provides IDE and command-line visibility into AI model token use, latency, spending, and forecasts, while consolidating monitoring across multiple models and connecting local data to organization-wide dashboards and budget controls.
Oct 28, 2026
915 words in the original blog post.
Amid a surge of critical October 2026 vulnerabilities described as “Black October,” the piece highlights PHP2Shell remote-code-execution flaws in PHP, a Node.js Brotli decoder memory-corruption issue called Node2Shell, and weak eMudhra certificate authority keys that leave Python’s certifi users exposed until updated. It argues that patching is increasingly difficult because vulnerabilities often reside in runtimes, base images, and legacy services rather than application code, while organizations must identify affected workloads, coordinate ownership, deploy updates across all regions, and confirm that no outdated instances remain active. The article presents New Relic Security RX as a runtime-focused complement to conventional scanners, using APM agent data to identify actively loaded vulnerable versions, prioritize findings, link them to service telemetry, and keep issues open until every running instance is remediated. It recommends prioritizing live production exposure, removing unused services, monitoring errors and suspicious runtime behavior during deployments, applying official upgrades rather than relying on mitigations, and tracking remediation status across both online and temporarily offline workloads.
Oct 28, 2026
2,854 words in the original blog post.
New Relic Infrastructure 360 is an advanced observability layer for platform engineers, SREs, and developers that combines multi-cloud resource discovery, dependency mapping, and configuration-change tracking to help diagnose outages and reduce mean time to resolution. Built on New Relic’s unified platform data model, it extends baseline infrastructure monitoring with progressive diagnostic capabilities, including health and alert context, topology and blast-radius visualization across AWS, Azure, Kubernetes, gateways, load balancers, and microservices, and property-level comparisons through its Configuration Explorer. The product is intended to reduce manual investigation across separate dashboards by correlating infrastructure changes with application performance issues and identifying instrumentation gaps. Its topology and change-context data also supports New Relic Autopilot by providing inputs intended to improve automated root-cause analysis and remediation recommendations. Infrastructure 360 was scheduled to enter public preview on October 13, 2026.
Oct 27, 2026
2,024 words in the original blog post.
New Relic has announced general availability of Oracle Cloud Infrastructure monitoring, bringing metrics, logs, tags, and metadata from more than 40 OCI services into its unified observability platform alongside AWS, Azure, GCP, application, and Kubernetes telemetry. The integration is intended to reduce fragmented monitoring and improve incident response for organizations operating large, multi-region, and multi-tenancy OCI environments by centralizing visibility into services such as Compute, Autonomous Database, Oracle Kubernetes Engine, and Networking. Setup uses OIDC Workload Identity Federation rather than static credentials, with guided UI, Terraform, and Oracle Resource Manager deployment options that provision OCI Connector Hub pipelines and enable telemetry collection across selected regions, namespaces, and log groups. Once data is ingested, users can access prebuilt Golden Metric dashboards, query telemetry through NRQL, and configure threshold-based or AI-driven anomaly alerts for metrics including CPU, memory, storage IOPS, and network throughput. OCI monitoring does not require a separate product SKU and is covered under New Relic’s standard usage-based pricing model.
Oct 27, 2026
1,113 words in the original blog post.
Forward deployed engineers (FDEs) are emerging as key partners for enterprises adopting AI, combining deep platform expertise with customer business knowledge to translate advanced technology into measurable outcomes. The article argues that effective AI agents require the same contextual access and onboarding as human employees, including connections to operational telemetry and systems of record, while continuous evaluations should be built around each agent’s intended business goal and updated as models, applications, and connectors change. It presents AI adoption as an ongoing organizational evolution involving people, processes, and autonomous systems rather than a one-time technology decision. Citing increased AI-generated code and production incidents, the piece emphasizes observability as essential for ensuring agents are accurate, reliable, cost-efficient, and aligned with business policies. New Relic positions its AI observability offerings, including evaluations, prompt operations, Ground Truth, and Autopilot SRE agents, as tools for monitoring agent behavior, identifying regressions, maintaining human oversight, and demonstrating value after FDE engagements conclude.
Oct 27, 2026
1,486 words in the original blog post.
AI-assisted development is increasing code volume and placing greater pressure on code review, deployment, and production verification processes, prompting engineering teams to seek runtime safeguards alongside traditional static checks. A New Relic and LaunchDarkly discussion describes combining LaunchDarkly’s feature flags and rollout controls with New Relic’s real-time telemetry to create a closed operational loop in which teams release changes gradually, observe telemetry, evaluate performance against predefined guardrails, and automatically disable problematic features when thresholds are exceeded. OpenTelemetry can attach feature-flag context to distributed traces, helping teams identify whether specific flag states are associated with errors or latency during complex rollouts. The approach aims to reduce manual incident triage, shorten customer impact, and allow continuous deployment while limiting exposure, with recommended early adoption focused on critical services, trace enrichment, and clearly defined rollback criteria.
Oct 26, 2026
958 words in the original blog post.
Datadog and Grafana differ primarily in deployment responsibility, pricing behavior, and the degree of control organizations retain over their observability stack. Datadog is a fully managed SaaS platform with integrated infrastructure monitoring, logs, traces, APM, and Watchdog AI analysis, minimizing operational work but potentially producing less predictable costs as custom metrics and telemetry cardinality rise. Grafana can be self-hosted through components such as Loki, Tempo, and Mimir or consumed through Grafana Cloud, offering flexibility and open-source control but requiring either substantial platform engineering effort or cloud usage spending; its pricing can also increase with active metric series. Both platforms provide anomaly detection and AI-assisted investigation, though Datadog presents these capabilities as an integrated platform layer while Grafana combines them with a composable ecosystem. The comparison argues that organizations often seek to reduce both SaaS cost uncertainty and self-hosting overhead, presenting New Relic as a managed alternative that supports OpenTelemetry, Prometheus, and Grafana integration while pricing primarily by data ingestion and consolidating telemetry analysis in one platform.
Oct 23, 2026
1,813 words in the original blog post.
Datadog and Splunk are often used together because Datadog specializes in cloud infrastructure monitoring, application performance monitoring, distributed tracing, and real user monitoring, while Splunk is widely used for large-scale log search, SIEM, compliance, threat detection, and environments with legacy or residency requirements. Their complementary roles can create operational friction during incidents, as teams may need to manually correlate metrics, traces, and logs across separate data models, alerting systems, and interfaces. Both platforms use pricing models that can rise substantially with growth, with Datadog combining per-host charges and usage-based fees for services such as logs and APM, and Splunk offering ingest-, activity-, or workload-based pricing. The post argues that maintaining both tools can lead to duplicated costs and slower troubleshooting, citing broader industry research on observability-tool sprawl. It presents New Relic as a consolidated alternative for infrastructure monitoring, APM, and log management through a shared data model and usage-based pricing, while acknowledging that it does not replace Splunk’s specialized SIEM and compliance capabilities.
Oct 23, 2026
2,050 words in the original blog post.
Platform engineering is presented as a combination of five complementary tool categories rather than a single product: developer portals and service catalogs, platform orchestration, infrastructure as code, CI/CD and GitOps, and observability. The discussion compares representative tools in each area, including Backstage, Port, OpsLevel, Cortex, Spacelift, Crossplane, Humanitec, Terraform, OpenTofu, Argo CD, Jenkins, Qovery, and New Relic, emphasizing that their suitability depends on organizational needs, existing investments, licensing preferences, Kubernetes expertise, and maintenance capacity. Effective internal developer platforms aim to reduce developers’ cognitive load by providing discoverable services, governed self-service infrastructure, repeatable deployments, and consistent visibility into system health without creating additional dashboards or rigid workflows that teams bypass. The text notes that observability frequently predates formal platform initiatives and can connect provisioning, deployment, and troubleshooting through standardized telemetry, logs, traces, and alerts, particularly when based on open standards such as OpenTelemetry. It recommends beginning with the category that creates the greatest current operational pain or manual ticket volume, accounting for implementation costs and temporary performance tradeoffs while ensuring that tools integrate into a coherent, widely adopted paved road.
Oct 22, 2026
2,172 words in the original blog post.
Network performance monitoring tools differ primarily between specialists that provide deep network telemetry and unified observability platforms that relate network issues to applications and infrastructure. Key evaluation factors include protocol and telemetry coverage, hybrid and multicloud support, root-cause investigation speed, deployment model, scalability, and total ownership costs. The comparison profiles New Relic for application-impact correlation; Kentik for flow analytics, capacity planning, and DDoS mitigation; LogicMonitor for broad hybrid IT monitoring; Cisco ThousandEyes for external internet and SaaS path visibility; SolarWinds NPM for self-hosted, on-prem network management; ManageEngine OpManager for device-based mid-market monitoring; Auvik for automated topology and MSP management; PRTG for flexible sensor-based all-in-one monitoring; and Zabbix for customizable open-source deployments. It argues that dedicated NPM products are most appropriate for protocol-level troubleshooting and network engineering, while unified platforms can accelerate incident response where the main concern is how network health affects services and customer experience; many mature organizations use both approaches.
Oct 22, 2026
2,075 words in the original blog post.
Datadog and Dynatrace differ primarily in architecture: Datadog offers a modular suite of separately licensed monitoring products that teams configure selectively, while Dynatrace uses its OneAgent and Smartscape topology model to automatically discover and map monitored environments. Datadog’s approach provides flexibility but can require more integrations, configuration, and product-specific spending, whereas Dynatrace enables rapid initial visibility but may create governance complexity at enterprise scale and depends on what OneAgent can instrument. Their AI tools, Watchdog and Davis AI, respectively analyze anomalies and causal dependencies, but both are constrained by the telemetry and products within their coverage. Both platforms use usage-based pricing that can become difficult to forecast during growth or traffic spikes, through Datadog’s multiple product meters or Dynatrace’s consumption against an annual commitment. The piece argues that both can also contribute to tool sprawl and context switching during incidents, then presents New Relic as an alternative built around a unified data model, OpenTelemetry support, cross-data correlation, and ingestion-based pricing; these comparisons and recommendations are published by New Relic and therefore reflect its perspective.
Oct 22, 2026
2,074 words in the original blog post.
New Relic is expanding its Business Development and Sales Development teams in Midtown Atlanta, positioning the office as a hub for early-career sales talent under Chief Revenue Officer Hein Hellemons. Located near Technology Square and Atlanta’s cultural, dining, and recreational attractions, the office emphasizes an energetic collaborative environment where representatives can share strategies, receive immediate coaching, and develop professionally. The company presents BDR and SDR roles as pathways to Account Executive positions, offering experience engaging engineering and IT leaders, using AI-enabled sales and CRM tools, attending community and industry events, and developing enterprise-sales skills. Employee testimonials highlight a supportive, low-ego culture, leadership investment in development, and opportunities to work across territories and participate in in-person events.
Oct 22, 2026
993 words in the original blog post.
New Relic Control is an observability control plane designed to reduce the operational burden of managing agents and telemetry across hybrid infrastructure, Kubernetes, and multi-cloud environments. It combines Fleet Control for centralized fleet management and automated deployments, Agent Control for remote orchestration and lifecycle monitoring using the Open Agent Management Protocol, and Pipeline Control for filtering, sampling, transforming, redacting, and enriching telemetry before ingestion. Linux and Windows host fleet support is now generally available, extending existing Kubernetes capabilities and enabling teams to standardize configurations, automate upgrades through release channels and canary rollouts, and manage integrations such as Apache, MySQL, and Flex. The platform supports New Relic agents, NRDOT collectors, Prometheus, and Fluent Bit, while fine-grained access controls, approval workflows, and audit logs support organizational governance. Fleet Control and Agent Control are included for New Relic users, whereas Pipeline Control is offered through the Advanced Compute SKU.
Oct 20, 2026
1,051 words in the original blog post.
New Relic has introduced public-beta support for natively receiving OpenTelemetry traces from Claude Managed Agents, the Messages API, and Claude SDKs, allowing Claude model and agent activity to appear alongside application performance data in a single distributed trace. By configuring New Relic as an OpenTelemetry destination in the Claude Console and exporting application traces to New Relic, teams can connect model requests, agent turns, tool calls, evaluations, and custom service activity without duplicating span data or maintaining custom attribute mappings. New Relic AI Observability can display trace waterfalls, agent entities, conversation replays, alerts, and dashboards, helping users identify whether delays or failures originate in models, tools, APIs, or databases. Its AI Evaluation feature, currently in public preview, can screen captured prompts, completions, and tool activity for prompt injection, PII leakage, and toxicity when full-content capture is enabled, while a metadata-only option limits collection to timing and identifiers. Initial support covers first-party Claude Platform services, with Amazon Bedrock and Google Vertex AI support planned, and organizations can configure regional data ingestion for selected workspaces through the Claude Console.
Oct 14, 2026
853 words in the original blog post.
New Relic has announced the general availability of Database 360, an observability platform that connects application performance monitoring transactions, relational database queries, execution plans, wait states, locks, and infrastructure metrics in a unified view. Designed to help developers, SREs, and database teams identify whether performance problems stem from application code, query design, database contention, or host resources, the product aims to reduce investigative delays and cross-team friction. Database 360 uses OpenTelemetry-based collection through NRDOT or CNCF-supported libraries, positioning it as an alternative to proprietary database-monitoring agents and separate point solutions. It also provides visibility into long-running sessions, blocking queries, resource-intensive workloads, client application behavior, and potentially oversized or underused managed database instances. The release includes new integrations for MySQL 8.0.3+ and PostgreSQL 14+, plus expanded capabilities for Oracle 19c+ and Microsoft SQL Server 2017+, with features such as query-plan correlation, session tracking, and curated root-cause-analysis workflows.
Oct 13, 2026
929 words in the original blog post.
New Relic Ground Truth connects AI coding agents to live operational data through the open Model Context Protocol (MCP), giving developers access to telemetry, logs, traces, deployments, alerts, and metrics without leaving their editor or terminal. It translates natural-language requests into NRQL queries, helping developers investigate errors, assess deployment health, examine latency, and identify production regressions without manually searching dashboards. Examples include Claude Code generating deployment health reports and beginning fixes, Kiro using production CPU and memory data to recommend Infrastructure as Code right-sizing changes, and GitHub Copilot detecting missing observability coverage, creating remediation issues, and implementing fixes through pull requests. New Relic also describes custom GitHub agents for incident analysis and emphasizes that its MCP Server works across supported AI environments rather than being tied to a single vendor. The company positions this production context as a way to reduce context switching, speed root-cause analysis and incident resolution, and improve developer productivity, with OAuth or API-key authentication and setup documentation available for adoption.
Oct 13, 2026
1,665 words in the original blog post.
As powerful AI models become widely available and easier to replace, the passage argues that durable enterprise advantage will increasingly come from operational context rather than model capability alone. In software operations, AI needs more than telemetry to make reliable recommendations; it also requires current, trustworthy knowledge of service dependencies, deployments, ownership, incident history, business priorities, and organizational decisions. This gap between what AI can technically do and what organizations are willing to trust it to do is described as the AI Trust Gap, which can be narrowed by building shared operational understanding and maintaining governance and human accountability. The passage presents observability as evolving from a way to collect and view metrics, logs, traces, and events into a foundation for trusted operational intelligence that helps engineers and AI interpret events together. It portrays Autonomous Operations not as replacing people, but as progressively expanding AI assistance while preserving human judgment, with New Relic’s Ground Truth and Autopilot cited as tools intended to support this approach.
Oct 13, 2026
2,081 words in the original blog post.
As software environments grow across applications, cloud infrastructure, Kubernetes, databases, and services, organizations face increasing difficulty establishing and maintaining meaningful alert coverage without expanding manual operational work. New Relic Smart Alerts uses historical telemetry to recommend alert conditions, coverage, and monitoring practices, giving engineers a more informed starting point while retaining their control over thresholds and monitoring strategy. The approach aims to improve detection rather than automate incident response, helping reduce gaps in monitoring and limit low-value alert noise that can weaken later investigation efforts. The piece presents effective detection as a necessary foundation for autonomous operations, where observability data, operational context, AI-assisted investigation, workflows, governance, and human oversight work together throughout the incident lifecycle.
Oct 13, 2026
1,295 words in the original blog post.
Modern reliability extends beyond uptime to encompass customer experience, business impact, rapid failure detection, and effective mitigation, particularly for distributed systems where technically available services can still perform poorly. New Relic describes improving its platform reliability by more than 50% through continuous measurement of time to detect, time to mitigate, customer impact, and change failure rates, with incident metadata correlated against broader operational telemetry. Its resilience strategy includes formal failure-mode reviews, cell-based and active-active architectures, multi-CDN failover that reportedly reduced CDN-related customer impact by 60%, and automated workload migration when infrastructure degradation signals emerge. To reduce detection time, the company uses automated alert-coverage checks, Terraform-based alerting standards, and end-to-end synthetic tests that validate complete customer data flows; these efforts reportedly improved time to detect by 40% in a year. For faster mitigation, it combines change tracking with AI-supported incident investigation and recommended actions, aiming to automate routine responses and reserve engineers for issues requiring human judgment.
Oct 09, 2026
2,089 words in the original blog post.
As cloud-native architectures, microservices, Kubernetes, and AI make production systems more complex, observability is evolving from the collection and visualization of telemetry into a source of trusted operational intelligence. Metrics, logs, and traces provide visibility, but effective incident response requires connecting those signals with dependencies, topology, ownership, deployment history, and institutional knowledge to explain what is happening and why. The article argues that this shared context is especially important as AI becomes an active collaborator in software operations, since AI recommendations depend on the completeness and reliability of the operational information available to them. New Relic positions Ground Truth as a capability for unifying telemetry and operational context, while Autopilot is intended to help engineers investigate, evaluate responses, and coordinate operational work using agentic AI. This proposed model of Autonomous Operations emphasizes that AI should support rather than replace human judgment, with organizations retaining control over approvals, governance, accountability, security, and business decisions.
Oct 06, 2026
2,233 words in the original blog post.
New Relic introduces AI Evaluation, a capability within its AI Observability platform designed to address semantic and safety failures in production generative AI systems that traditional monitoring may not detect. Using asynchronous “LLM-as-a-judge” models, it evaluates sampled prompts and completions for criteria such as accuracy, relevance, hallucinations, prompt injection, PII leakage, toxicity, bias, and RAG faithfulness, assigning normalized scores, pass/fail results, and explanatory reasoning. These results are attached directly to distributed traces, allowing teams to connect low-quality AI outputs with specific prompts, retrieval calls, backend services, infrastructure behavior, and token costs. The service also supports configurable sampling to limit evaluation expense while maintaining oversight, and is positioned for AI/ML engineers, platform and operations teams, and security and compliance personnel seeking to troubleshoot AI failures, compare model cost versus quality, and apply production guardrails without separate monitoring silos.
Oct 06, 2026
1,553 words in the original blog post.
New Relic argues that AI is accelerating software development while increasing production complexity, creating a need for connected and trusted operational context across applications, infrastructure, AI models, agents, and teams. Its strategy centers on AI Observability, which tracks AI coding-tool use, model and agent behavior, response quality, cost, latency, drift, and security risks; Autonomous Operations, which combines telemetry, changes, dependencies, and organizational knowledge to help teams investigate incidents while retaining human control over actions; and Platform Engineering, which uses live telemetry to maintain service catalogs, ownership information, infrastructure maps, standards scorecards, and configuration-change visibility. Together, these capabilities aim to replace fragmented inventories and manual evidence gathering with a shared intelligence foundation that helps organizations understand what is running, how systems connect, what has changed, and how AI can be operated more reliably as automation expands.
Oct 06, 2026
1,752 words in the original blog post.
New Relic AI Observability for OpenTelemetry is presented as a platform for unifying open-source OTel and native APM telemetry to address fragmented visibility into generative AI applications, token spending, performance, and security risks. Its “Normalize on Read” approach keeps OTel data in native span tables and dynamically adapts queries at runtime, aiming to avoid data duplication, ingestion transformations, and associated storage and processing costs. The platform also supports account-wide model inventories and cross-instrumentation comparisons of cost, latency, errors, and quality, enabling teams to assess different LLM configurations. To identify “Shadow AI,” it scans incoming traces for GenAI semantic attributes and automatically tags discovered AI-enabled services, creating a current inventory without manual registration. It further routes OTel trace data into evaluation pipelines for measures such as PII masking, prompt-injection detection, and toxicity screening. New Relic states that organizations can use existing OpenTelemetry instrumentation and trace routing without custom code changes or vendor-specific SDKs.
Oct 06, 2026
1,495 words in the original blog post.
LaunchDarkly and New Relic are introducing an integration intended to connect feature-flag-controlled releases with full-stack observability, addressing the difficulty of tracking how rapidly deployed changes affect production systems. LaunchDarkly manages gradual, targeted rollouts and rollbacks, while New Relic supplies real-time metrics such as latency, errors, traces, and performance baselines. The combined workflow allows teams to begin releases with limited user exposure, evaluate telemetry during rollout, and automatically pause or revert a feature when predefined thresholds are exceeded, limiting the impact of regressions. The integration also adds feature-flag context to logs and traces, helping engineering, SRE, platform, product, and leadership teams identify affected user cohorts and isolate the source of incidents more quickly. The capability is presented as a limited preview, with an interest form available for teams seeking to test it.
Oct 06, 2026
880 words in the original blog post.
New Relic’s Advanced APM bundle is presented as an intelligence-led observability offering intended to help enterprise engineering, SRE, and DevOps teams move from reactive incident response to proactive performance management. It combines Transaction 360, which automatically maps end-to-end dependencies across applications, databases, infrastructure, and frontend systems, with Intelligent Workloads, which maintains dynamic business entities as cloud environments change, and custom business KPI tracking that connects technical degradation to measures such as checkout duration, abandoned carts, and revenue risk. Its Performance Risks Inbox continuously identifies coding and database anti-patterns, including N+1 queries and slow SQL, before they trigger alerts or outages, while providing remediation guidance. The platform also uses nested health rollups and longer-term transaction-flow retention to help teams identify hidden failures, analyze degradation over time, and reduce mean time to resolution. New Relic positions these capabilities as particularly relevant for ephemeral cloud architectures and AI-driven workloads, arguing that they can reduce diagnostic effort, protect service-level agreements, and align reliability work with business outcomes.
Oct 06, 2026
1,430 words in the original blog post.
New Relic Pathpoint is presented as a business observability application that connects end-to-end digital business journeys with the underlying health of distributed technical systems, helping organizations assess both operational failures and their business consequences during incidents. It visually maps workflows such as browsing, checkout, and payment into stages, associates each stage with APM, infrastructure, and synthetic-monitoring signals, and displays business KPIs including order volume, abandonment, and revenue impact. Dual-condition health rules aim to distinguish technical alerts from KPI declines caused by nontechnical factors, while journey-oriented views are intended to help responders identify the affected stage and direct the appropriate engineering teams more quickly. The platform also gives executives and business stakeholders self-service visibility into process health and financial exposure, reducing reliance on engineers to translate telemetry during outages. Pathpoint supports flexible flow modeling, templates, collaborative drafts, query libraries, historical incident playback, and Terraform and NerdGraph API integrations for enterprise-scale governance and infrastructure-as-code deployment.
Oct 06, 2026
1,278 words in the original blog post.
New Relic Smart Alerts is presented as a telemetry-driven approach to reducing the growing operational burden of creating and maintaining alert coverage across dynamic cloud environments. Rather than automatically changing thresholds or replacing engineering judgment, it analyzes historical system behavior to recommend relevant alert conditions, coverage, and monitoring practices, helping teams apply established standards more consistently across large numbers of services and infrastructure entities. The approach aims to improve the quality of detection without merely increasing alert volume, balancing comprehensive monitoring with reduced noise and manual configuration work. Smart Alerts also supports New Relic’s broader Autonomous Operations strategy by strengthening the early detection layer that AI-assisted investigation and operational workflows can use, while illustrating a broader shift in observability from collecting evidence about past events to learning from telemetry to improve future operational decisions.
Oct 06, 2026
1,350 words in the original blog post.
AI agents require specialized observability because their non-deterministic, multi-step behavior can produce failures such as stalled tool calls, hallucinations, loops, and failed handoffs that conventional request-based monitoring cannot adequately explain. The overview compares eight platforms across multi-agent tracing, evaluation and quality scoring, integration with existing APM, infrastructure, and log telemetry, and deployment options: New Relic and Datadog extend full-stack observability platforms with agent tracing; Langfuse, LangSmith, Arize, Braintrust, Comet Opik, and Confident AI emphasize varying combinations of open-source or self-hosted telemetry, prompt management, automated and human evaluation, regression testing, and security checks. It distinguishes visibility gaps, where teams cannot determine what an agent did during an incident, from evaluation gaps, where teams can inspect traces but cannot reliably assess declining answer quality, hallucinations, bias, or prompt regressions. Selection should be based on testing real multi-agent handoffs, ensuring scoring methods are calibrated and supported by human review, correlating agent activity with underlying systems, and confirming deployment requirements such as VPC or self-hosted support. Teams already using a broad observability platform may benefit from extending it for agent telemetry, while organizations focused on prompt quality, edge-case testing, and red-teaming may need a dedicated AI-native evaluation platform, with many ultimately requiring both capabilities.
Oct 05, 2026
2,556 words in the original blog post.
Incident management platforms span two complementary functions: detecting and diagnosing problems through observability and AIOps, and coordinating response through on-call schedules, escalation policies, collaboration, status updates, and postmortems. The comparison assesses nine common options according to alerting depth, collaboration support, monitoring integrations, and pricing, noting that faster resolution depends on both reliable upstream detection and effective downstream coordination. New Relic emphasizes telemetry-based correlation and root-cause analysis, PagerDuty offers mature and configurable paging, and Opsgenie is being retired in favor of Jira Service Management by April 2027. Incident.io and Rootly focus on chat-based coordination, ServiceNow supports formal ITIL and compliance workflows, FireHydrant automates response runbooks, Jira Service Management consolidates Atlassian-native service processes, and Better Stack combines monitoring, status pages, and on-call tooling for smaller teams. Key capabilities include flexible escalation paths, alert deduplication and correlation, automated incident records and postmortems, and integrations with monitoring systems, with organizations often combining specialized detection and coordination tools rather than relying on a single platform.
Oct 05, 2026
2,302 words in the original blog post.
Organizations evaluating alternatives to PagerDuty are encouraged to weigh alerting reliability, escalation flexibility, pricing predictability, and the breadth of incident-response capabilities as alert volume and burnout increase. The options profiled include incident.io for chat-native coordination, Jira Service Management as Atlassian absorbs Opsgenie functionality, Rootly for automation-focused growing teams, Better Stack for bundled monitoring and on-call tools, FireHydrant for service-oriented incident orchestration, SolarWinds IT Incident Response following Squadcast’s acquisition, and Grafana Cloud IRM following the sunset of self-hosted Grafana OnCall. The discussion emphasizes testing realistic schedules, missed-page escalations, service-based routing, future total costs, and the quality of monitoring integrations rather than relying on similar-looking feature lists. It also argues that changing paging platforms alone cannot solve alert fatigue, citing outage and productivity figures to support the value of observability tools such as New Relic, which positions itself as an upstream detection layer that correlates alerts, identifies likely root causes, and updates incident records before unnecessary notifications reach on-call staff.
Oct 02, 2026
2,306 words in the original blog post.
FinOps tools vary primarily between platforms centered on financial management, such as allocation, chargeback, forecasting, governance, and Reserved Instance or Savings Plan automation, and observability platforms that connect cost changes to deployments, services, queries, and infrastructure telemetry. The comparison highlights New Relic and Datadog for correlating cloud, Kubernetes, and AI costs with operational performance data; CloudZero for unit economics; Finout for virtual tagging and multi-cloud allocation; Vantage for developer-led optimization and commitment automation; IBM Cloudability for enterprise financial governance; and CloudHealth for combining cost control with security and compliance policies. Key evaluation factors include cost allocation, anomaly detection, operational correlation, commitment management, multi-cloud support, and increasingly AI cost visibility, as organizations seek to manage AI spending alongside conventional cloud use. The appropriate tool depends on an organization’s immediate needs, though mature teams may combine a dedicated FinOps platform for financial accountability and rate optimization with an observability platform for investigating the technical causes of cost anomalies.
Oct 02, 2026
2,304 words in the original blog post.