Launched this week
Progress AI Observability
Trace, evaluate, and improve AI agents in production
234 followers
Trace, evaluate, and improve AI agents in production
234 followers
Debug and monitor AI agent failures in minutes. Trace every run, catch hallucinations and ungrounded answers that traditional monitoring misses, and see exactly what went wrong. Reduce token waste, improve agent quality, and ship faster with support forNET, Python, and JavaScript.
Hey Product Hunt 👋
I’m Lyubo, the Product Manager behind Progress AI Observability.
We know a lot of teams are building AI agents now. The first demo comes together quickly but then you try to run it in production and things get more complicated.
Traditional monitoring can tell you that an app is running, but it usually can't explain why an agent chose a particular tool, ignored useful context, entered an expensive loop or produced an answer that looked convincing but was wrong.
We built Progress AI Observability to give engineering teams that missing visibility.
We want to move teams from “something went wrong” to understanding why it happened, what needs to change and whether the next version is actually better.
You can start tracing in minutes with support for .NET, Python, and JavaScript/TypeScript. There’s a free plan, and no credit card is required.
We'd love to hear your feedback on:
What’s hardest to debug once an agent reaches production?
Which signals are most useful to you: traces, evaluations, latency, token usage, or cost?
What would you need to see before using this with a production agent?
Thanks for checking it out 🙏
- Lyubo and the Progress AI Observability Team
Progress AI Observability
@lyuatanasov 🙏🙏🙏
Progress AI Observability
@lyuatanasov soo excited for the ProductHunt community to check us out, hope this can be super useful to anyone building agents!
Progress AI Observability
@lyuatanasov Really excited to finally have this out 🚀🚀🚀 Really looking forward to the feedback from the community!
@lyuatanasov 🔥🔥🔥
PicWish
@lyuatanasov getting setup in under 5 mins with 10k free units is ideal for prototyping. do those units get consumed faster by LLM as Judge evaluation runs or raw trace spans?
Hey Mohsin Ali - 1 evaluation is indeed 2 units, so they get consumed just a bit faster. From what I've seen with existing devs, 10k seems to be a good volume for POCs and prototyping. By the way, if over time we notice it's not sufficient, we'll definitely bump it up - we want to encourage people to experiment and build more cool stuff with LLMs.
PromptWave AI
Great to hear you see the value in our solution, Kris! Full observability + evaluations are indeed fundamental for any AI agent out in the wild.
I assume if I want to monitor how Claude Code works, then I have to use Fiddler? (I love Fiddler)
Hey, Jay @jay_janarthanan1 you can absolutely also capture the traces from Claude Code in Progress AI Observability (we support OTLP) . Of course fiddler works perfectly fine, if you don't have to peristently store the traces . Dropping you some instructions. If you can't get it working DM me and I will try to help you out !
HarnessRouter
@yochev Congratulations. And happy product launch.
Progress AI Observability
@huisong_li TY mate!