1. X
  2. langfuse.com
Log inSign up
langfuse.com
942 posts
Image
user avatar
langfuse.com
@langfuse
Open source AI engineering platform. Now part of @clickhousedb. We're hiring: langfuse.com/join-us
github.com/langfuse/langfuse
langfuse.com
Joined July 2023
670
Following
5,386
Followers
1
Subscription
AffiliatesAffiliatesRepliesRepliesMediaMedia

New to X?

Sign up now to get your own personalized timeline!

Create account

By signing up, you agree to the Terms of Service and Privacy Policy, including Cookie Use.

Terms·Privacy·Cookies·Accessibility·Ads Info·© 2026 X Corp.
Don't miss what's happening
People on X are the first to know.
Log inSign up
  • Pinned
    user avatar
    langfuse.com
    @langfuse
    May 14
    Building high-quality AI systems is hard. At Langfuse we see the best AI teams converging on a process to get complex AI systems to production. We call it the AI Engineering Loop. Check out the first piece of our series and find out more in our academy
    user avatar
    Annabell
    langfuse.com
    @annabellschfr
    May 14
    Article cover image
    Article
    The AI Engineering Loop - Building quality AI Systems that scale
    This is one piece of a series we’re publishing as part of the Langfuse Academy, where we walk through the full AI engineering lifecycle. The AI Engineering Loop Building high-quality AI systems that...
    306K0306K
  • user avatar
    langfuse.com
    @langfuse
    Jul 16
    langfuse now has a bug bounty. as part of @ClickHouseDB, langfuse is in scope for the ClickHouse @Bugcrowd program. report vulnerabilities there and get rewarded for valid findings. scope + focus areas:
    Image
    Responsible Disclosure - Langfuse
    From langfuse.com
    1.8K01.8K
  • user avatar
    langfuse.com
    @langfuse
    Jul 15
    Improvement Loops can boost quality or cut cost - either way, you need a clear evaluation set (a 'target function') to self-improve on. We ran one on a classification task and shared what it taught us about human expertise, at @aiDotEngineer youtu.be/eAXxdtNlK04
    Image
    user avatar
    Annabell
    langfuse.com
    @annabellschfr
    Jul 14
    Coding agents always had one target function to loop to: code_compiles == true Most other fields need to find ways to bake domain expertise into their 'target function' first to loop anywhere. I made the case for domain expertise in datasets bootstrapping your agent performance
    1.3K01.3K
  • user avatar
    langfuse.com
    @langfuse
    Jul 13
    Stop shipping prompts on vibes. With Langfuse you can test AI features on real datasets, run prompt versions side-by-side, and auto-score them with evals. This way you catch regressions before your users do.
    Image
    00:00
    2.6K02.6K
  • user avatar
    langfuse.com
    @langfuse
    Jul 8
    meet us at @WeAreDevs in Berlin, 08-10 July booth 2_41. come talk agents, evals, and continuously improving them.
    Image
    1.1K01.1K
  • user avatar
    langfuse.com
    @langfuse
    Jul 7
    AI now in the langfuse filter search bar. describe filters in plain language, "enterprise customers stuck in the membership-support queue", and it drafts editable pills mapped to your project's actual columns and metadata keys. langfuse.com/changelog/2026…
    Image
    00:00
    2.7K02.7K
  • user avatar
    langfuse.com
    @langfuse
    Jul 6
    langfuse datasets are now multi-modal. attach images, audio, video, and documents to dataset items via the UI or python/js sdk, then run multi-modal experiments. media auto-resolves to bytes, base64, or data uris for any provider. langfuse.com/changelog/2026…
    Image
    2.2K02.2K
  • user avatar
    langfuse.com
    @langfuse
    Jul 2
    This AI agent confidently gave a customer the wrong answer. It took 30 seconds in Langfuse to find why: one bad line in a routing table, buried three steps upstream from where the mistake actually showed up. Video breakdown below 👇
    Image
    00:00
    1.5K01.5K
  • user avatar
    langfuse.com
    @langfuse
    Jun 30
    Your OpenClaw agent is doing a lot behind the scenes, model calls, tool runs, costs. Now you can actually see all of it. Trace every step with the official Langfuse plugin, just a couple of commands to set up 👇
    Image
    00:00
    1.2K01.2K
    user avatar
    langfuse.com
    @langfuse
    Jul 1
    You can visit our Docs on the integration here: langfuse.com/integrations/o…. And our ClawHub Plugin is here: clawhub.ai/codecls/plugin… 1. CLI: openclaw plugins install clawhub:openclaw-x-langfuse-plugin 2. Edit the openclaw.json file to add the keys and region 3. CLI: openclaw gateway
    Image
    Trace OpenClaw with Langfuse - Langfuse
    From langfuse.com
    3520352
  • langfuse.com reposted
    user avatar
    Marc Klingen
    langfuse.com
    @marcklingen
    Jun 29
    Langfuse workshop ongoing at @aiDotEngineer on how to auto-optimize AI Agents If you can’t join live, check out what @annabellschfr and @lotte_verheyden built at langfuse.com/workshop
    Image
    1.1K01.1K
  • langfuse.com reposted
    user avatar
    Annabell
    langfuse.com
    @annabellschfr
    Jun 26
    Just arrived to SF for @aiDotEngineer and ran into our posters at Powell street🎉🪢 Gave the pic to AI to make the agents look as happy as I was!
    Image
    00:00
    Image
    1.1K01.1K
  • user avatar
    langfuse.com
    @langfuse
    Jun 26
    code evaluators means running untrusted user code across thousands of tenants in our multi-tenant cloud. @wochinge wrote about the full design: (V8 isolates, pyodide), gVisor and firecracker, and kubernetes jobs.
    Image
    Designing the runtime for Langfuse code evaluators - Langfuse
    From langfuse.com
    1.2K01.2K
  • langfuse.com reposted
    user avatar
    Nikita Kabardin
    langfuse.com
    @nkabardin
    Jun 23
    My work schedule for today. Empty calendar, full tunnel mode. If you also want suspiciously calm but shipping-intense days like this, join @langfuse
    Image
    1K01K
  • user avatar
    langfuse.com
    @langfuse
    Jun 22
    web callouts are live. trigger an HTTP call to your own backend from any trace, observation, or session. re-run an agent, attach a trace to a support case, or kick off your own workflow. langfuse.com/docs/observabi…
    Image
    00:00
    7600760
Advertisement
Advertisement