Tag
Dunetrace, an open-source real-time monitoring tool for production AI agents, updates with cross-agent pattern analysis, Langfuse deep analysis integration, and custom agent support.
The article highlights that agent harnesses cause a 30-50 point performance swing compared to model selection, arguing that teams should focus on instance-level verification rather than just model names.
Fleet agents now support configurable tracing per agent, allowing developers to enable or disable detailed trace information for better debugging.
A promotional post for Honeycomb Innovation Week 2026, a free 3-day virtual event (May 12-14) focused on observability in the agent era, featuring keynotes, product launches, and partnerships.
The official TanStack AI OpenTelemetry support is now available, offering an open-source backend for traces, datasets, and replay to improve debuggability.
A developer shares their experience of a single system prompt change degrading LLM response quality without triggering traditional monitoring alerts, and describes internal tooling they built to monitor semantic quality in production LLM applications.
This article introduces 7 production-ready skills from the Hermes Skills Hub, covering the full lifecycle from tool integration and structured output to deployment, observability, and security.
A detailed breakdown of a 9-layer production AI architecture covering RAG pipeline, agents, prompts, security, evaluation, and observability layers.
The article discusses the challenges of debugging and evaluating LLM judges using Arize Phoenix, which traces evaluator runs via OpenTelemetry to inspect decision logic, costs, and potential biases.
Tracea is a new product offering Datadog-like observability for AI agents, providing features such as tracing, root cause analysis, and team memory.
Noah at Smart Rent coins "Telemetry-Driven Development" for Elixir: instrument first with OpenTelemetry, then ship, replacing guess-work with production data from 848k Nerves gateways.
A curated list of 11 links shared daily to help people learn AI evaluation techniques, covering evals, observability, LLM-as-judge, and agent evaluation.
VictoriaMetrics presented retroactive sampling at KubeCon EU 2026, a new method that significantly reduces traffic, CPU, and memory overhead compared to traditional tail sampling in OpenTelemetry pipelines.
Datadog integrated OpenAI's Codex into their code review process and found it detected 22% of historical incidents that human reviewers missed, demonstrating superior system-level reasoning capabilities compared to traditional static analysis tools.
White Circle raised $11M to launch a unified AI control platform offering red-teaming, guardrails, observability, and optimization for enterprise deployments.