langfuse.com

Open source AI engineering platform. Now part of @clickhousedb. We're hiring:

[langfuse.com @langfuse]

@marcklingen sat down to demo all langfuse features in a single video.

Agent observability, monitoring, evals, prompt management, and what is special about Langfuse.

Great place to start if you are new, or catch up on latest changes.

https://t.co/3lSorqdNme

[langfuse.com @langfuse]

@nimarblu @ClickHouseDB @iamleowolters @hassiebpakzad @maxdeichmann @LysanderKie @ceo_clickhouse 🏃‍♂️

[langfuse.com @langfuse]

RT @marliessophie: We 100x our team in 2 years.

> Then: @langfuse offsite 2024 with 7 people
> Now: @ClickHouseDB offsite 2026 with 800 p…

[langfuse.com @langfuse]

@marliessophie @ClickHouseDB 🚀

[langfuse.com @langfuse]

@nimarblu on the thinking behind Halo

[Nimar @nimarblu]

I'm a strong believer in the power of laziness, especially for engineers. Laziness as a driver often forces good solutions.

You go into the Langfuse UI to achieve something, eg to find the important traces to look at, setup an evaluator to be alerted when something breaks or optimize a prompt.
Locally, we often did this via Agent + CLI/MCP, but cumbersome.

Therefore, we built Langfuse Halo 😇👼 to watch over your traces in-app and proactively help you solve problems
1/x

[langfuse.com @langfuse]

@spacexai “List this week’s most expensive Cursor turns and what they have in common.”

Ask Cursor, update your instructions, then compare cost, latency, and task completion.

Read the guide: https://t.co/PoE6FYcsjV

[langfuse.com @langfuse]

Two Langfuse plugins are now available on the @spacexai Marketplace:
→ 𝘓𝘢𝘯𝘨𝘧𝘶𝘴𝘦 𝘖𝘣𝘴𝘦𝘳𝘷𝘢𝘣𝘪𝘭𝘪𝘵𝘺 captures your agents’ model calls, tool executions, and token usage.
→ 𝘓𝘢𝘯𝘨𝘧𝘶𝘴𝘦 gives your agents skills to investigate traces, manage prompts, and run evals.

Analytics powered by @ClickHouseDB

[Boris Michel @drborismichel]

Ask Halo challenge

The brief: find out why eligible customer applications were getting rejected.

In seven minutes, @langfuse's Halo:

- found an ignored €6,000 subsidy in traces
- tested a revised prompt and
- made it live, fixing the application workflow

Here's the demo.

[langfuse.com @langfuse]

Since Halo uses the Langfuse skill, it's very good at helping you with workflows like cost optimization. Give it a try 😇

[Lotte @lotte_verheyden]

Pro tip: based on this guide, Halo can analyze your traces and give you suggestions on how to optimize your agent costs

[langfuse.com @langfuse]

Another great example of how Halo saves you time 😇

[Lotte @lotte_verheyden]

Setting up dashboards got a lot more efficient with Halo!

I haven't created a dashboard manually in a while, and we're seeing the same with our users: almost 10% of Halo conversations involve creating dashboards

[langfuse.com @langfuse]

@catmanyau exactly

[langfuse.com @langfuse]

New article on how we built Halo 😇

A couple of decisions:
- Halo uses the same MCP server, skill and docs as external agents accessing Langfuse, so every new Halo functionality is immediately available to everyone.
- The agent loop runs separate from its sandbox.
- We use Langfuse to observe and improve Halo.

Read the full article here: https://t.co/mv6mJ6WMRf

[langfuse.com @langfuse]

@felix_muller_ai We know our users 😇 most of them prefer their own coding agent, so making that experience great is important to us

[langfuse.com @langfuse]

experiments now show which metric moved, and why.

chart quality, cost, and latency across runs. compare each score against your baseline. filter to items under a score threshold to find the regressions, then annotate them from the keyboard.

https://t.co/MXBtiQuVQh

[langfuse.com @langfuse]

😇😇😇

[Lotte @lotte_verheyden]

Here you see Halo help me get started with evals. It looks at my project, determines that the best first step is capturing user feedback signals, and proposes a list of signals I could capture.
Really helpful when starting out, Halo can be a great sparring partner in getting to a good observability setup

[langfuse.com @langfuse]

@marcklingen 😇

[Subquadratic @subquadratic]

SubQ Snap is now available in @langfuse.

Model ID: subq-snap-preview

https://t.co/UmbD86ZmKj

[Lotte @lotte_verheyden]

Here you see Halo help me get started with evals. It looks at my project, determines that the best first step is capturing user feedback signals, and proposes a list of signals I could capture.
Really helpful when starting out, Halo can be a great sparring partner in getting to a good observability setup

[langfuse.com @langfuse]

@benedictk__ @marcklingen @ClickHouseDB 🏃‍♂️

[langfuse.com @langfuse]

@Shubhamgilada it's treated as trace content, not as a prompt!

[langfuse.com @langfuse]

meet Halo, the assistant that lives inside Langfuse 👀

ask about your traces, sessions and metrics in plain language, or have it build a dataset or dashboard for you. it asks before changing anything.