Posthog

Pricing change detected for Posthog

Pricing updated for Posthog: - Free: allowance changed from 1 project, 1-year data retention, free tier limits to Usage capped at free tier limits - Pay-as-you-...

We put PostHog in Slack and now everyone's an engineer

PostHog released a Slack app in beta that allows users to tag @PostHog with requests to generate code, fix bugs, or analyze product data. The bot creates sandbo...

Evaluations

Prompt Management

PostHog Slack app

PostHog introduced a Slack app in beta that integrates its AI agent into Slack channels. The bot can answer product data questions, execute code tasks, and mana...

Tracing

PostHog introduced Tracing, a distributed tracing solution integrated with OpenTelemetry (OTLP) that allows users to capture and explore traces without vendor-s...

AI Observability

PostHog introduced AI Observability features to capture traces, generations, and spans from AI and LLM products. The update includes guides for integrating with...

How autoresearch found a 3-year-old bug in our query engine

PostHog’s AI agent discovered a three-year-old bug in their ClickHouse query engine where timestamp filters failed to use the primary key correctly due to a tim...

Minimum viable AI observability: what to set up after shipping your first AI feature

PostHog published a guide for small teams on implementing minimal AI observability, focusing on tracing, cost tracking, and error capture as foundational elemen...

Pricing change detected for Posthog

Pricing updated for Posthog: - Free: promotion ended (was: No credit card required) - Free: allowance changed from 1 project, 1-year data retention to 1 project...

Pricing change detected for Posthog

Pricing updated for Posthog: - Free: new promotion — No credit card required - Free: allowance changed from 1 project, 1-year data retention, free tier limits t...

Training our own AI models

PostHog announced plans to train AI models on customer data to power new proactive, self-driving features like session replay analysis and synthetic user testin...

From 270GB RAM to 5GB: Moving local flag evaluation from Django to Rust

PostHog migrated its feature flags local evaluation endpoint from a Django-based deployment (30 pods, 270GB RAM) to Rust, reducing CPU by 24x and memory by 56x ...

The best analytics tool stack for vibe-coded apps

PostHog published a guide advocating for early analytics setup in vibe-coded apps, highlighting product analytics, web analytics, and LLM observability as essen...

The do's and don'ts of minimum viable product marketing

PostHog’s product marketing team prioritizes tactical, iterative launches over rigid frameworks, emphasizing clear announcements, conflict-driven storytelling, ...

The best MCP servers for developers at startups

PostHog’s blog outlines six key workflows where Model Context Protocol (MCP) servers streamline startup operations, such as debugging, shipping code, and analyz...