Log inSign up
Log inSign up
Comet
3,507 posts
Comet profile banner
@Cometml

Comet

@Cometml
Comet provides an end-to-end model evaluation platform for AI developers, with best in class LLM evaluations, experiment tracking, and production monitoring
New York, NY
comet.com
Joined October 2017
878 Following
14.9K Followers
RepliesRepliesRepostsRepostsMediaMedia

Log in or sign up for X

See what’s happening and join the conversation

Continue with phone
or
Log in with username or email
Terms·Privacy·Cookies·Accessibility·Ads Info·© 2026 X Corp.
  • @Cometml
    Comet
    @Cometml
    Oct 1
    Cutting AI costs isn’t just about routing to cheaper models. Coding agents can waste tokens in plenty of other places. Our CEO @gidim unpacks what we learned building Cost Intelligence in a @cerebral_valley deep dive.
    10x Engineer or 10x Token Bill? 💸
    From cerebralvalley.beehiiv.com
    6
  • @Cometml
    Comet
    @Cometml
    Sep 30
    We tested #Jev against gpt-4o-mini as the judge model evaluating the same set of 1,000 application traces. Jev came out 3.5x cheaper and 3.8x faster than the LLM judge, and agreed with it on 897 of them. Here's how we set up the experiment and what we learned:
    Jev vs. LLM-as-a-Judge for AI Evals
    From comet.com
    5
  • @Cometml
    Comet
    @Cometml
    Sep 24
    We’re excited to announce that Opik is now available within @nebiusai Cloud Applications, giving AI builders application-level agent observability and evaluation inside one of the most powerful end-to-end AI cloud platforms available today. Learn more about using Opik within
    Comet Extends Agent Observability and Evaluation to Nebius AI Cloud
    From comet.com
    6
  • @Cometml
    Comet
    @Cometml
    Sep 22
    What if each #ClaudeCode session contained distinct, identifiable phases, and those phases could be used to optimize usage patterns and token cost without hurting productivity? Our team explores this concept as a new feature for Comet Cost Intelligence:
    Capturing a 400-Turn Claude Code Session in a Single Image
    From comet.com
    2
  • @Cometml
    Comet
    @Cometml
    Sep 16
    Traditional APM wasn’t built for LLMs. It assumes deterministic behavior and loud failures. That’s the observability gap our Principal Engineer, Andrés Cruz, has been tackling while building Opik from the early days 🧵
    3
Edit with