Releasesby IbisHow we built automatic clustering for LLM traces
GA7 months agoUnverifiedAdded Sep 24, 2026
PostHog has introduced automatic clustering for LLM traces and generations. This feature helps AI product teams analyze usage patterns and identify outlier behaviors in their AI applications.
Topics: LLMs, Observability
More PostHog releases
Every PostHog releasePostHog is now available in Vercel’s v0
PostHog has extended its Vercel integration to v0 via MCP. This allows v0 to generate code based on real product data, including experiment results and feature flag states.
PostHog × Vercel: feature flags, minus the plumbing
PostHog has released an integration with Vercel that syncs feature flags and experiments into Vercel's native Flags system, automating credential management.
Your logs' final destination is in GA. You always end up here anyway
PostHog Logs is now generally available, providing a centralized destination for logs alongside errors, session replays, and product data.
PostHog Code and the self-driving product
PostHog Desktop, formerly PostHog Code, is now in beta. It is a desktop application that runs coding agents on top of product data to automate bug fixes, UX issues, and conversion tweaks.
Training our own AI models
PostHog has announced plans to train models on internal data to improve proactive features like session replay analysis and synthetic user testing.
We put PostHog in Slack and now everyone's an engineer
The PostHog Slack app is now in beta, allowing users to trigger automated bug fixes and feature development directly from Slack threads.
What if your product built itself?
PostHog has released a pipeline that ingests product signals, uses research agents to identify actionable problems, and automatically opens GitHub pull requests.
What is a Scout? A technical deep dive
PostHog has introduced scouts, which are autonomous agents that monitor product data for patterns and draft fixes. This feature is now available in open beta.
Also shipped on Mar 11, 2026
PostHog in March 2026Introducing Replit Agent 4: Built for Creativity
Replit Agent 4 is now available, featuring an infinite canvas for design, parallel agent execution for backend and frontend tasks, and support for mobile and web app development.
Purchase and manage additional AI credits
Rolling out this week, admins can now purchase additional AI credits to use in Figma.
eBPF Network Metrics reach GA with enhanced APM capabilities
As we move through 2026, the landscape of software engineering is accelerating at an unprecedented pace. We’re excited to share several updates across the New Relic eBPF platform, including the General Availability of New Relic eBPF Network Metrics, enhancements to eBPF APM. New Relic eBPF Network Metrics is now GA!
Sender constrained tokens using DPoP is now Generally Available on Enterprise plans
Support for sender constraining tokens using Demonstrating Proof of Possession (DPoP) is now generally available on Enterprise plans. Demonstrating Proof of Possession (DPoP) as defined in RFC9449, is an application level mechanism for binding tokens issued by Auth0 to the client application that requested that token.
Secrets Management UI available in Public Preview
Secrets management launched in Grafana Cloud in August 2025 , with integrations in Synthetic Monitoring, and later expanded to Grafana Cloud k6 workflows and Git Sync. Now, we’re introducing a centralized Secrets Management UI that brings secret operations into one place across Grafana solutions.
Cloudflare Fundamentals - JSON responses and RFC 9457 support for Cloudflare 1xxx errors
Cloudflare-generated 1xxx errors now return structured JSON when clients send Accept: application/json or Accept: application/problem+json . JSON responses follow RFC 9457 (Problem Details for HTTP APIs) ↗ , so any HTTP client that understands Problem Details can parse the base members without Cloudflare-specific code.