AI observability can cost more than it should. We ran the numbers: same scale, same workload, four tools. At moderate production load, Logfire is up to 40x less expensive than the alternatives. Spans-based pricing. No proprietary units. AI native and full-stack observability in one platform. See the full comparison: pydantic.dev/articles/ai-obs…

At OpenAI, we're continuing to bet on Rust as the future of systems programming. I'm proud to announce that we're making a $600,000 commitment to the Rust Foundation, which combines our Platinum membership with additional support for maintainer efforts across the Rust ecosystem.

Your chatbot is throwing 500s since Tuesday's deploy, but which pod, on which node, after which release? The new Kubernetes view in Pydantic Logfire takes you from the failing pod to the trace that explains it in one click. Read the post: ordnl.link/dzeRb7v

Your agent workflow times out on requests. The model is fine. So the slow thing is one of the 14 services it's calling. But which one? The new Services view in Pydantic Logfire takes you from the failing service to the exact trace that explains it: ordnl.link/PIog5LE

Your AI was fine last week. New prompt, new model, answers quietly got worse. Nothing errored. Sound familiar? Join us next Thu, Jun 25 for an online workshop on why AI observability should cover both the LLM layer and the whole system around it. Register: ordnl.link/no3DsOS

big shoutout to all of the sponsors making this possible @pydantic @nvidia @CrusoeAI @AnthropicAI @OptiverGlobal and massive kudos to @zqwq333 and team for getting all of this together ❤️

Your refund agent returns the wrong answer. No crash, no error, no stack trace. The bug is in the seam between multiple observability tools, and none sees the whole run. Pydantic Logfire puts the agent and infra views in one product: ordnl.link/ewPYoKB

excited for @northflank to be supporting the hottest hackathon happening in london rn!! alongside @dawncapital @pydantic @nvidia @CrusoeAI @AnthropicAI, and more hot quant summer kicks off today 🥵

Building developer tools (more precisely, my fun in life, reusable UI components) in the agentic era feels like a totally different discipline. The perfect tool must be practically invisible for the engineer. The agent installs, configures, and integrates it in your environment without making a fuss. The agent should silently translate what the engineer specifies to the concepts of the tool. Nobody cares about your elegant API shapes. Human-readable docs are increasingly irrelevant, maybe still matter for discoverability. Marketing/demos is kind of ok, as they will help for agent consumption. The API ergonomics/intuitiveness now mean different things, as there are no longer fingers pressing keyboard keys. The agent feels no pain producing large pieces of code, as long as it works and performs well. Error reporting, correct tracing, and observability now matter much more (Hello @pydantic Logfire), as the agent needs to be able to easily figure out what's wrong with its own slop. Performance diagnostics are very valuable. And, a bit sad, but I can't escape the feeling that I'm building some invisible infrastructure that will no longer have direct human interaction. But hey, at least we can make the end users life a bit better.

AI spend is the new headcount.

New: Team and Growth orgs on Logfire now get a one-time $10 starter credit for Pydantic AI Gateway. Lands automatically when you upgrade. Nothing to claim, no code to paste.

actually it's pronounced @pydantic