# 8 learnings from 1 year of agents - PostHog AI

DevFeed: [8 learnings from 1 year of agents - PostHog AI](<https://devfeed.tech/articles/8-learnings-from-1-year-of-agents-posthog-ai-80450.md>)

Original publisher: [Read original article](<https://posthog.com/blog/8-learnings-from-1-year-of-agents-posthog-ai>)

Author: Michael Matloka

Published: 2025-11-25T00:00:00Z

Content type: article

Language: en

Sources: [PostHog](<https://devfeed.tech/sources/posthog.md>)

Topics: [agent orchestration](<https://devfeed.tech/topics/agent-orchestration.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [human in the loop agent](<https://devfeed.tech/topics/human-in-the-loop-agent.md>), [LLM observability](<https://devfeed.tech/topics/llm-observability.md>)

Tags: [agentic](<https://devfeed.tech/tags/agentic.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-architecture](<https://devfeed.tech/tags/ai-architecture.md>), [ai-observability](<https://devfeed.tech/tags/ai-observability.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [code](<https://devfeed.tech/tags/code.md>), [context](<https://devfeed.tech/tags/context.md>), [learnings](<https://devfeed.tech/tags/learnings.md>)

## AI overview

PostHog describes lessons from a year of building its AI product agent. The article discusses how model improvements, a simple looping architecture, retained task context, visible tool activity, and reviews of real user traces shaped the agent. It also explains tradeoffs in using orchestration frameworks and evals.

## Source excerpt

Today we launch PostHog AI, the AI agent built into PostHog . A year in the making, we've gone a long way from our first chat prototype made over a...