"Distributed teams, equipped with modern AI tools, have structural advantages that in-person teams don't." Our CTO David Montague on why the remote work debate has shifted. A few things that used to be weaknesses of remote work, and aren't anymore: - Meetings that vanished: Ad hoc huddles now get transcribed in one click and turned into PRDs by Claude. Try doing that in a conference room. - Invisible contributions: AI can summarize a quarter's worth of someone's work in a way that's fairer than lines-of-code or story points ever were. - Private offices for everyone: Focus, voice dictation without self-consciousness, space to think. Perks that used to be reserved for execs. The research backs this up. A randomized controlled trial at Trip.com, published in Nature in 2024, followed 1,612 employees over two years and found hybrid/remote arrangements reduced quit rates by 33% with no measurable impact on performance or promotions. Full arguments in the comments. If this resonates, we're hiring at Pydantic.
Pydantic
Software Development
The Pydantic Stack: Build with AI at scale, without fail with Pydantic Logfire, Pydantic AI, Pydantic Evals & AI Gateway
About us
End-to-end AI engineering stack We started as a Python validation library. We're now the AI engineering company behind the stack that teams use to build with GenAI in production. Pydantic AI. Pydantic Logfire. Pydantic Evals. AI Gateway. Each tool is useful on its own. Together, they cover the full lifecycle of building with AI: from structured outputs and agent logic, to observability, evaluation, and cost tracking. Trusted by developers building at scale. Developer experience first, always. Pydantic, because AI is still just engineering.
- Website
-
https://pydantic.dev
External link for Pydantic
- Industry
- Software Development
- Company size
- 11-50 employees
- Headquarters
- California
- Type
- Privately Held
- Founded
- 2022
- Specialties
- Observability, AI Agents, AI workflows, FinOps, and Traces and metrics
Locations
-
Primary
Get directions
California, US
Employees at Pydantic
Updates
-
Vstorm just published an independent comparison of the top 5 observability tools for agentic AI systems in production. Logfire came out on top for Python-first teams. When an AI feature is slow, you want to see the full picture. Database query, model call, post-processing, in one trace. Most tools only show you one of those. The article is worth reading if you're choosing an observability stack right now. It covers LangSmith, Langfuse, Arize Phoenix, and Datadog alongside Logfire, with honest tradeoffs for each. Link in comments.
-
-
We built a honeypot. We're inviting you to break it. Monty is our sandboxed Python interpreter, built to run LLM-generated code safely. We think it holds. So we put a secret on the machine and opened it up. POST Python to the endpoint, escape the sandbox, find the secret, and we'll pay you a bounty. Read the rules before you start (link in comments). Then, go ahead and break things.
I may live to regret this 😱, but here goes: Announcing the Pydantic Monty bounty program: if you can escape the sandbox and read secrets on the host, we'll pay $5,000. Full deals in the blog post https://lnkd.in/d3naG3QQ and at https://hackmonty.com.
-
Pydantic reposted this
Your agent needs a sandbox, not a desert. That one line from Samuel Colvin, CEO and Founder of Pydantic, tells you everything about where most agent deployments go wrong. If you're building agents in production, this is the session you don't want to miss at AgentCon Silicon Valley. Monday, 4 May 2026 at the Computer History Museum, Mountain View. Free to attend. 👉 https://lnkd.in/gp6bDvxx #AgentCon #GlobalAICommunity #AIAgents #AgentConSiliconValley #Pydantic
-
-
Our newest blogpost is a practical guide to optimizing MCP server tool definitions for reduced token usage, using snapshot testing in Rust. https://lnkd.in/e9QwZMqc #mcp #tokenefficiency #servertool #tutorial
-
Pydantic reposted this
The Power of Rust in Python 🦀🐍 I'm really excited about this month's main talk at PyData Southampton. I first saw David Hewitt at the Pydantic @ Google London Meetup last July. David gave a fantastic demo using an agentic AI solution to monitor the BlueSky firehose, showcasing Pydantic Logfire for observability. One of the core technical takeaways was how he used Rust to get around performance bottlenecks in the data stream. Fast forward to now, and I am thrilled that we are bringing David to PyData Southampton on Tuesday, 21 April! I am eager to hear how I can start bringing Rust into my own workflows. David is giving us a deep dive into "Getting Started with PyO3" so we can learn how to leverage the power of Rust in our own code. If you are building data-heavy apps and want to see how the "engine" behind Pydantic and Polars actually works, you cannot miss this one. See you at Carnival House at 19:00! 🍕🚀 #PyData #Pydantic #RustLang #Python #DataEngineering #DataScience #Performance #PyO3 #SouthamptonTech #OpenSource #SoftwareEngineeringT
🦀 Python Performance, Powered by Rust: A Pydantic Special PyData Southampton 🐍 The Python ecosystem is evolving. While we love the simplicity of Python, the modern performance heavyweights we use daily, like Polars and Pydantic, owe their incredible speed to a "native core" written in Rust. We are thrilled to announce that for our next meetup, we are welcoming David Hewitt, Staff Data Engineer at Pydantic and a lead maintainer of PyO3! 🎤 The Talk: Getting Started with PyO3 - David will take us under the hood of how Rust and Python work together. Whether you are building high-throughput data pipelines or looking to optimise your local libraries, this session is a masterclass in modern performance engineering. You will learn: ✅ Practical steps to integrate Rust into your Python projects using PyO3. ✅ Real-world examples of Rust code running seamlessly inside Python. ⚡ Call for Lightning Talks! Got a 5-minute project, a tool you have built, or a "lesson learned" to share? We have slots open! Reach out to us now to grab a spot on the stage. 🍕 The Perks: As always, we will have plenty of pizza, refreshments, and the legendary PyData networking. Plus, two lucky attendees will win a personal 1-year JetBrains license! 🗓 When: Tuesday, 21 April 📍 Where: Carnival House, Southampton 🎟 Secure your spot here: See link in comments Don't miss the chance to learn from a world-class engineer shaping the tools we use every day. See you there! #PyData #Pydantic #RustLang #Python #DataEngineering #PyO3 #SouthamptonTech #OpenSource #SoftwareEngineering #Polars #PerformanceOptimisation
-
Our founder, Samuel Colvin at no10.
Absolute honour to be invited to 10 Downing Street by David Gelberg today to talk about AI and Pydantic. Surreal that the UK government is in a large part defined in Pydantic models. I continue to be impressed by how the UK government is adopting and leveraging AI. Now we just need a British model!
-
-
We contributed to Apache Arrow making Parquet bloom filters smaller and more effective. Full write up by Adrian Garcia Badaracco on what it means for Pydantic Logfire and the wider Parquet ecosystem. Link in comments.
-
-
Vercel's AI SDK emits rich OpenTelemetry spans for every LLM call: full prompts, responses, token counts, streaming latency, and tool call details. All following the OpenTelemetry GenAI semantic conventions. Logfire speaks the same standard. Point your OTLP exporter at Logfire and you get conversation replay, token tracking, cost calculation, and streaming performance metrics, no custom integration code. The setup is three steps: two environment variables, a one-liner instrumentation file, and one option on your AI SDK calls. Petyo Ivanov wrote up how it works on our latest blogpost. https://lnkd.in/ebTiXsHi
-
Pydantic reposted this
We Just Published a New Guide: Emerging AI Agent Frameworks Developers Should Watch in 2026 (Part 2) After our previous deep dive into AI agent frameworks, we’re back with Part 2 — exploring the next wave of tools redefining how intelligent systems are built. The first wave of AI agent frameworks solved the basics — prompt chaining, tool calling, and simple workflows. But the game has changed. We’re now entering a new era where frameworks aren’t just tools… They’re redefining how intelligent systems are built. In Part 2, we break down 5 powerful frameworks shaping this shift: ⚡ Pydantic AI → Type-safe, production-ready agent systems ⚡ VoltAgent → Full-stack platform for building + operating agents ⚡ Google ADK → Code-first, modular agent engineering ⚡ OpenAI Agents SDK → Lightweight, composable multi-agent systems ⚡ Haystack → Context-first pipelines for real-world AI If you're building anything serious in AI right now, this guide will change how you think about agents. Read the full guide here: https://lnkd.in/gCD_Wy5D #aiagents #opensource #aiframeworks
-
