Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
17 commits
Select commit Hold shift + click to select a range
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
108 changes: 108 additions & 0 deletions contents/blog/self-driving-ga.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,108 @@
---
title: "Self-Driving is out of beta (we finally fixed the pricing, too)"
date: 2026-11-09
rootPage: /blog
sidebar: Blog
showTitle: true
hideAnchor: true
featuredImage: https://res.cloudinary.com/dmukukwp6/image/upload/Self_driving_pricing_1x_d2d6846f08.png
featuredImageType: full
author:
- joe-martin
category: PostHog news
tags:
- Product updates
- AI
seo:
metaTitle: "Self-Driving is out of beta (we finally fixed the pricing, too)"
metaDescription: "PostHog Self-Driving is out of beta. See how Juno, SoundBoost, Higgsfield, and our own team use it, and how the new per-run pricing works."
---

In July we launched PostHog [Self-Driving](/self-driving) as an open beta. Our agents dug through your product data, found things worth fixing, and tried to fix them. Their fix rate is now high enough that it's out of beta, and it has real pricing.

If pricing is all you came for, skip ahead to the pricing section. For everyone else, here's what thousands of teams, including ours, are doing with it.

## Juno merged 26 of the last 30 PRs our agents opened

[Juno](/customers/juno) is an AI health assistant for people living with chronic illness. It came out of YC's Spring 2026 batch and already has 250,000 users, but it's a hard product to debug. It has to strip PII everywhere, it runs on multiple platforms, its users vary wildly, and the team behind it is small.

Self-Driving doesn't need to see the personal data, and it doesn't care how many platforms or engineers you have. It pieces the story together from Error Tracking and Product Analytics, writes it up as a [report](/docs/self-driving/reports), and turns actionable reports into PRs if you want them.

Juno's team merged 26 of the agents' last 30 PRs straight away. Some fixed bugs, like chat boxes that hung. Others added features, like a "Log all" button that makes medication tracking quicker for people juggling several prescriptions. All of them made health tracking a bit less tiring for Juno's users.

## SoundBoost rolled back a domain migration because a report told it to

[SoundBoost](/customers/soundboost) is an AI mastering tool with 225,000 users and a team of four. Founder Berkan Cesur starts every morning in the Self-Driving Inbox, where reports rank what to fix next. "If we fix 10% here, 5% there? Those gains add up, and PostHog compiles that into a menu for me," he says.

Most mornings, reports are bug fixes and performance work, but one morning it was an existential problem. Self-Driving noticed daily clicks from Google had dropped to zero shortly after SoundBoost moved from an .ai domain to a .com.

The team assumed Google was lagging on the new domain and was ready to leave it be, but the report showed that conversions had stopped, traffic was down, and their best funnel was gone. They rolled back to .ai immediately and the Google traffic came back.

The diagnosis was fast because Berkan tracks every button and feature as an [action](/docs/data/actions), which he can then query using the [PostHog MCP](/mcp) in his editor. He used to review every self-driving PR line by line. Now he merges them almost instantly.

## Our self-driving team let Self-Driving drive itself

We run Self-Driving on PostHog itself, because obviously. One [report](/docs/self-driving/reports) showed that our agents passed on thousands of chances to open a PR each day, and no chart showed why. The agent behind it wrote a query, broke the skips down by reason, checked the numbers, and added a chart to the team dashboard.

![A self-driving report on why our agents skipped PRs](https://res.cloudinary.com/dmukukwp6/image/upload/q_auto,f_auto/Screenshot_2026_11_03_at_10_05_19_27f74b08da.png)

The chart got <TeamMember name="Andy Maguire" />, a product engineer on the team, digging into why the work was failing. That investigation is a big part of why Self-Driving is out of beta today. It also taught us that a report can be worth as much as a PR.

Another report compared merge rates across the models our agents can use to write PRs. It flagged that one model had roughly half the merge rate of the others, then Self-Driving kept watching to check the gap held as the sample grew. It held. The model was GPT-6-luna-flex, by the way.

We can't fix the model, but we can steer users away from it before it gives them a bad experience, so that's what we do now. Without that report we might not have noticed for weeks.

<iframe
width="560"
height="315"
src="https://www.youtube-nocookie.com/embed/QefP1WAXhc0"
title="See what self-driving looks like in practice"
frameborder="0"
allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture"
allowfullscreen
></iframe>
<Caption>See what self-driving looks like in practice. Warning: This video includes pickleball</Caption>

## Higgsfield merges self-driving PRs without opening PostHog

Self-driving PRs go straight to GitHub, and the Inbox is accessible over MCP and other surfaces. If your team works AI-native and rarely logs in to PostHog's web interface, it still fits your workflow.

[Higgsfield](https://higgsfield.ai/), an AI creative suite, works this way. The team has merged over 120 self-driving PRs, all of them from GitHub. It points Self-Driving at error tracking [signals](/docs/self-driving/signals) and reviews the output alongside their daily work. Fixes so far include broken Stripe tracking and noisy exceptions that showed up as errors when they weren't.

## "All I want to know about is the pricing"

Every new org gets $10 off Self-Driving in its first month. That's enough for up to five reports, so you can see what the agents find in your product before you pay anything. After that, you pay for what you use:

- **You pay per run.** A scout scan, a signals investigation, or a PR implementation each counts as one run, and each one tells you what it cost.
Comment thread
joethreepwood marked this conversation as resolved.
Resolved

- **Not all runs cost the same.** Cost depends on how much context the agent reads and how long it spends in a sandbox. Light runs cost cents. Deep dives into your data cost more.

- **You only pay for what you turn on.** Choose reports, PRs, or both. Billing limits cap what you spend.

In the beta we charged $15 per PR and gave the reports away for free. That turned out to be backwards. The research is where the agents do most of the work, and it's the part people care about most, so now every run is billed, reports included. Automatic PRs are turned off by default, so you can keep an eye on how useful they are while you get used to the system.

> **Controlling your costs:** We've updated [our Self-Driving pricing docs](/docs/self-driving/pricing#ask-your-agent) with more information, including a [`managing-self-driving-spend` skill](https://github.com/PostHog/posthog/blob/master/products/signals/skills/managing-self-driving-spend/SKILL.md) you can call through [the PostHog MCP](/mcp).

### If you used the beta

Your org stays on beta pricing until December 9, 2026. From that day you move to per-run pricing like everyone else.

For some of you that means a bigger bill. We'll email every org whose monthly cost goes up by more than $200, or more than doubles past $50 (about 300 of you), and offer a call with one of our engineers to tune your setup, so you're not paying for work that isn't earning its keep.

<!-- TODO: restore this sentence at the end of the paragraph above once PostHog/posthog#124825 is merged:
We've also built a skill to help you manage spend. Ask your agent to run the `managing-self-driving-spend` skill through the [PostHog MCP](/mcp) to get started. -->
Comment on lines +92 to +93

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@annikaschmid Are we still planning to have this skill for launch?


Two more things to know:

1. If you removed your billing limit during the beta, we're putting one back at $150. Raise it whenever you like. We'd rather you chose to spend the money than found out you had.

2. If you have credits or a billing limit, your spend might not change, but your usage might count against it at a different rate. If it's faster, you'll hit the limit earlier than you did in the beta and runs will stop. The call above is for you too.

## What's next for the Self-Driving Inbox?

We assumed auto-generated PRs would be the draw, and they are interesting, but once people had the beta in their hands they spent their time on the _reports_. We think that's because people care more about work they were involved in than work they didn't ask for. Better reports make better PRs anyway, if PRs are what you want.

You can see what the agents are working on right now. [Our docs have a public feed of real reports we found useful](/docs/self-driving/from-our-inbox), curated by our own agents. Once a day they read our Inbox, pick new examples, and open a PR to add them to the page.

Everything above is available today. [Set up Self-Driving](/docs/self-driving/setup), browse [examples](/docs/self-driving/scout-examples) of what else self-driving can watch, and check [what your first month looks like](/docs/self-driving/pricing).

2 changes: 1 addition & 1 deletion contents/blog/what-is-a-scout.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -20,7 +20,7 @@ seo:
metaDescription: 'A scout is a small scheduled agent that watches your PostHog data, learns what is worth knowing, and emits useful signals you (or your agents) can act on. Here is how they work, with real examples.'
---

A scout is a small agent (don't overthink that word) that looks at, wrangles, and makes sense of any data available in PostHog. It's a core component of [Self-driving](/docs/self-driving) in PostHog (now in open beta), [an autonomous loop](/newsletter/agent-autonomy#level-3-self-driving-mode) that monitors your product for patterns and drafts fixes.
A scout is a small agent (don't overthink that word) that looks at, wrangles, and makes sense of any data available in PostHog. It's a core component of [Self-driving](/docs/self-driving) in PostHog, [an autonomous loop](/newsletter/agent-autonomy#level-3-self-driving-mode) that monitors your product for patterns and drafts fixes.

A scout's job is to emit useful **signals** – things worth knowing – that you or your agents can act on. Sometimes that action happens automatically, like an agent opening a PR to fix what the scout found, sometimes it needs some human judgement before you and your agent can tackle it together.

Expand Down
2 changes: 0 additions & 2 deletions contents/compare/best-statsig-alternatives.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -80,8 +80,6 @@ Pair it with feature flags to roll out new models gradually, experiments to A/B

It ships with PostHog skills so the agent already knows your stack, runs tasks in cloud sandboxes so they keep going after you close your laptop, and opens GitHub PRs you can review without leaving the app.

[Currently in beta](/blog/self-driving-product).

</details>

For more details, here are our in-depth comparisons of the best Statsig alternatives:
Expand Down
2 changes: 1 addition & 1 deletion contents/compare/posthog-vs-sentry.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -289,7 +289,7 @@ Both PostHog and Sentry work with other tools, but PostHog can act more as a sin
Both companies are building on the idea that an agent, not a person, should do the first pass on an error.

- **Sentry** has Seer, a paid add-on ($40 per active contributor per month, not available on the free plan) that scans issues, finds root causes, and can open fix PRs, and a hosted MCP server that lets Claude, Cursor, and other clients query your Sentry data.
- **PostHog** has [PostHog AI](/ai) for asking questions and generating fixes, [Scouts](/blog/what-is-a-scout) that watch your product data and send reports to an [Inbox](/docs/self-driving/inbox) where agents pick them up and open PRs (in beta, three PRs a month free), a free [MCP server](/mcp) with a read-only mode, and a [Slack app](/slack) you can tag in a thread to ask a question or ship a fix.
- **PostHog** has [PostHog AI](/ai) for asking questions and generating fixes, [Scouts](/blog/what-is-a-scout) that watch your product data and send reports to an [Inbox](/docs/self-driving/inbox) where agents pick them up and open PRs, a free [MCP server](/mcp) with a read-only mode, and a [Slack app](/slack) you can tag in a thread to ask a question or ship a fix.

The difference is scope. Seer works from the error outward. PostHog's agents work from everything the product knows about the user, the session, the flags they saw, and the error, which is the same context argument that runs through this whole comparison.

Expand Down
6 changes: 0 additions & 6 deletions contents/docs/ai-engineering/ai-wizard.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -56,12 +56,6 @@ Running the wizard automatically:

## Self-driving

<CalloutBox icon="IconInfo" title="Open beta" type="fyi">

Self-driving is currently in open beta.

</CalloutBox>

The wizard is also how you turn on [self-driving](/docs/self-driving). A self-driving product can prompt itself: PostHog turns your product data into signals, and agents act on those signals to ship improvements – all inside the guardrails you set.

Once PostHog is installed and capturing events, run the self-driving setup. It enables your signal sources, sets up your scouts, and hands you a link to your inbox:
Expand Down
6 changes: 0 additions & 6 deletions contents/docs/ai-observability/self-driving.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -5,12 +5,6 @@ title: Self-driving with AI Observability
import { CallToAction } from "components/CallToAction";
import CalloutBox from 'components/Docs/CalloutBox'

<CalloutBox icon="IconFlask" title="Open beta" type="action">

Self-driving is in open beta. It's improving quickly – expect rough edges, and expect them to disappear fast.

</CalloutBox>

To power [Self-driving](/docs/self-driving), AI Observability turns your LLM data into [signals](/docs/self-driving/signals): agents investigate what changed, and reports land in your [inbox](/docs/self-driving/inbox) – where one click turns a finding into a pull request.

All of it needs LLM data flowing first – if you haven't instrumented yet, start with [installation](/docs/ai-observability/installation).
Expand Down
6 changes: 0 additions & 6 deletions contents/docs/self-driving/anatomy-of-a-pr.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -7,12 +7,6 @@ showTitle: true

import { CallToAction } from 'components/CallToAction'

<CalloutBox icon="IconFlask" title="Open beta" type="action">

Self-driving is in open beta. It's improving quickly – expect rough edges, and expect them to disappear fast.

</CalloutBox>

The concepts docs describe the [self-improving loop](/docs/self-driving/self-improving-loop) in the abstract. This page follows one bug all the way through it, from the first signal to the measured fix. The product is a composite example, but every step is exactly what self-driving does.

## 1. The signals arrive
Expand Down
6 changes: 0 additions & 6 deletions contents/docs/self-driving/context.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -7,12 +7,6 @@ showTitle: true

import { CallToAction } from 'components/CallToAction'

<CalloutBox icon="IconFlask" title="Open beta" type="action">

Self-driving is in open beta. It's improving quickly – expect rough edges, and expect them to disappear fast.

</CalloutBox>

Context is everything your agents need to know about your product, tied together in one place. We call it your **[context warehouse](/blog/what-is-a-context-warehouse)**, and it's what self-driving runs on.

Context helps your agents do their best work.
Expand Down
8 changes: 1 addition & 7 deletions contents/docs/self-driving/faq.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -7,12 +7,6 @@ showTitle: true

import WizardCommand from 'components/WizardCommand'

<CalloutBox icon="IconFlask" title="Open beta" type="action">

Self-driving is in open beta. It's improving quickly – expect rough edges, and expect them to disappear fast.

</CalloutBox>

Skeptical questions, answered straight. If yours isn't here, ask us via [in-app support](https://app.posthog.com/home#supportModal) or the [community](/questions).

## Won't this fill my repo with AI slop?
Expand Down Expand Up @@ -47,7 +41,7 @@ Yes, heavily. A troop of more than 90 scouts – the canonical fleet plus dozens

## What happens when it gets something wrong?

It will sometimes; it's in open beta and we say so on every page. The system is designed so that being wrong is cheap:
It will sometimes. The system is designed so that being wrong is cheap:

- A wrong report costs you nothing. Reports are free, and you can decline or snooze them.
- A wrong PR costs you a review, and then nothing: [we refund PRs](/docs/self-driving/pricing#refunds) that misdiagnose the problem or don't fix what they claim to.
Expand Down
6 changes: 0 additions & 6 deletions contents/docs/self-driving/from-our-inbox/index.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -7,12 +7,6 @@ showTitle: true

import FromOurInbox from 'components/SelfDrivingInbox/FromOurInbox'

<CalloutBox icon="IconFlask" title="Open beta" type="action">

Self-driving is in open beta. It's improving quickly – expect rough edges, and expect them to disappear fast.

</CalloutBox>

We run self-driving on PostHog itself. This page is a feed of real [reports](/docs/self-driving/reports) from our own inbox that turned into useful work. Nothing here is a mockup: each report was found by a [scout](/docs/self-driving/scouts) or a [signal source](/docs/self-driving/inbox/sources).

Not every useful report ends in code. Some end in a merged pull request, and we link to it. Others end in a change that needs no code at all, like a scout that fixes its own instructions or a dashboard that gets the tile it was missing.
Expand Down
6 changes: 0 additions & 6 deletions contents/docs/self-driving/inbox/implementation.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -7,12 +7,6 @@ showTitle: true

import { CallToAction } from 'components/CallToAction'

<CalloutBox icon="IconFlask" title="Open beta" type="action">

Self-driving is in open beta. It's improving quickly – expect rough edges, and expect them to disappear fast.

</CalloutBox>

Once [research](/docs/self-driving/inbox/research) is complete and a report is marked **Actionable**, an implementation agent builds the change and opens a pull request for you to review.

## Sandboxed execution
Expand Down
6 changes: 0 additions & 6 deletions contents/docs/self-driving/inbox/index.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -7,12 +7,6 @@ showTitle: true

import { CallToAction } from 'components/CallToAction'

<CalloutBox icon="IconFlask" title="Open beta" type="action">

Self-driving is in open beta. It's improving quickly – expect rough edges, and expect them to disappear fast.

</CalloutBox>

The inbox is where self-driving hands work back to you. [Reports](/docs/self-driving/reports) and the pull requests the loop opens both land here, ranked by priority, so the most important work rises to the top.

It's the human side of the [self-improving loop](/docs/self-driving/self-improving-loop): the loop watches, enriches, and proposes, and the inbox is where you review, steer, and decide.
Expand Down
6 changes: 0 additions & 6 deletions contents/docs/self-driving/inbox/research.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -7,12 +7,6 @@ showTitle: true

import { CallToAction } from 'components/CallToAction'

<CalloutBox icon="IconFlask" title="Open beta" type="action">

Self-driving is in open beta. It's improving quickly – expect rough edges, and expect them to disappear fast.

</CalloutBox>

When a [signal](/docs/self-driving/signals) arrives – an [error tracking](/docs/error-tracking) exception, a GitHub issue, anything from a connected source – an agent investigates it automatically. It uses the MCP servers in your workspace, starting with the PostHog MCP, to figure out what's happening and how big the impact is, with access to your product's code for grounding.

The result is a [report](/docs/self-driving/reports) rated on two axes:
Expand Down
6 changes: 0 additions & 6 deletions contents/docs/self-driving/inbox/sources.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -8,12 +8,6 @@ showTitle: true
import { CallToAction } from "components/CallToAction";
import List from "components/List";

<CalloutBox icon="IconFlask" title="Open beta" type="action">

Self-driving is in open beta. It's improving quickly – expect rough edges, and expect them to disappear fast.

</CalloutBox>

Signal sources are what fill your inbox. A [signal](/docs/self-driving/signals) can be a production error, a support conversation, a Session Replay pattern, a log alert changing state, a Replay Vision scanner finding, an LLM trace regression, an insight anomaly, or an issue from an external tracker. Each source watches one of these streams and turns what it finds into work worth investigating.

## Configure sources
Expand Down
6 changes: 0 additions & 6 deletions contents/docs/self-driving/inbox/troubleshooting.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -7,12 +7,6 @@ showTitle: true

import { CallToAction } from 'components/CallToAction'

<CalloutBox icon="IconFlask" title="Open beta" type="action">

Self-driving is in open beta. It's improving quickly – expect rough edges, and expect them to disappear fast.

</CalloutBox>

This page covers common issues with inbox reports and agent research.

## Reports are not appearing
Expand Down
Loading
Loading