Meta tags:
description=Pydantic is the AI engineering stack for teams building with AI. Observe with Logfire, evaluate with Pydantic Evals, build type-safe agents with Pydantic AI, and track LLM costs with AI Gateway.;

Headings (most frequently used words):

pydantic, ai, you, in, agent, the, agents, to, production, one, trace, your, logfire, on, frameworks, aibuild, ll, actually, ship, applications, fail, ways, existing, tools, can, see, across, model, calls, apis, and, database, gives, every, team, better, decision, brings, evidence, engineering, works, with, entire, stack, building, best, developer, experience, planet, earth, latest, articles, python, javascript, typescript, rust, any, language, via, opentelemetry, infrastructure, run, coding, e2b, sandboxes, runs, runson, code, is, graph, announcement, v2, 14, release, would, deploy, an, don, control, models, libraries, web, databases, dashboards, alerts, compute, metrics,

Text of the page (most frequently used words):
the (38), pydantic (25), and (23), logfire (22), agent (12), with (11), your (9), see (8), production (8), you (7), #engineering (7), agents (7), 2026 (6), can (6), opentelemetry (6), one (6), trace (6), model (6), across (5), evals (5), infrastructure (5), mins (5), october (5), what (5), evidence (5), from (5), for (4), stack (4), explore (4), validation (4), observability (4), monitoring (4), gateway (4), use (4), cases (4), behind (4), more (4), that (4), start (4), team (4), python (4), tool (4), application (4), cost (4), every (4), database (4), faster (4), sign (3), about (3), service (3), llm (3), enterprise (3), same (3), tools (3), they (3), data (3), not (3), apis (3), free (3), built (3), metrics (3), request (3), tracing (3), typescript (3), databases (3), frameworks (3), calls (3), without (3), context (3), systems (3), actually (3), into (3), failures (3), improve (3), how (3), services (2), wrong (2), newsletter (2), security (2), contact (2), github (2), support (2), all (2), monty (2), open (2), source (2), pricing (2), control (2), samuel (2), colvin (2), code (2), coding (2), articles (2), now (2), build (2), products (2), started (2), did (2), developer (2), problem (2), brings (2), building (2), our (2), people (2), cloud (2), hosts (2), postgresql (2), redis (2), keep (2), underneath (2), any (2), framework (2), works (2), sdks (2), are (2), ecosystem (2), rust (2), sdk (2), javascript (2), google (2), follow (2), fixes (2), path (2), than (2), budget (2), through (2), turn (2), eval (2), test (2), evaluation (2), together (2), api (2), real (2), healthy (2), caught (2), spike (2), queries (2), 150 (2), download (2), svg (2), end (2), trust, center, cookies, terms, privacy, inc, thanks, subscribing, something, went, please, try, again, subscribe, email, updates, dash0, elastic, honeycomb, signoz, grafana, sentry, datadog, arize, langsmith, langfuse, braintrust, alternatives, compare, careers, company, slack, documentation, events, customer, stories, blog, community, oss, levels, slos, frontend, app, performance, apm, governance, controls, judge, product, aditya, vardhan, would, deploy, don, victorien, plot, announcement, release, graph, bill, easton, runs, runson, laís, carvalho, run, e2b, sandboxes, latest, samuelcolvin, creator, growth, means, maintainers, get, principle, most, powerful, still, easy, working, out, frustration, type, hints, nothing, runtime, curiosity, whether, could, validate, work, business, idea, worth, solving, well, library, obsession, clear, useful, defaults, standards, whole, best, experience, planet, earth, mission, book, demo, dennis, griffin, sophos, tell, was, who, system, providers, aws, lambda, docker, kubernetes, compute, nginx, kafka, mongodb, elasticsearch, dashboards, alerts, clusters, containers, each, investigation, swift, erlang, elixir, php, ruby, net, java, instrumentation, convenient, wrappers, lock, boundary, language, via, log, crate, cloudflare, workers, deno, node, express, next, browser, vercel, mysql, pymongo, asyncpg, sqlalchemy, psycopg, wsgi, asgi, starlette, flask, django, fastapi, web, mcp, litellm, genai, anthropic, openai, models, libraries, matrix, mastra, strands, claude, adk, autogen, crewai, langgraph, verified, paths, tying, visibility, first, party, them, rebuilding, around, vendor, entire, integrations, decisions, touch, connects, queues, result, happened, inspect, enforce, policy, retain, sensitive, cross, expected, cut, earning, its, retries, choices, quietly, burn, layer, sprawl, turns, incidents, archaeology, verify, fix, synthetic, sets, miss, users, hit, move, symptom, explains, stitching, separate, guessing, averages, gives, better, decision, outcome, truth, keeps, full, execution, debug, broke, evaluate, worked, solution, timeout, elsewhere, missing, output, looks, answer, even, when, ignore, exercises, creative, writing, imagined, while, failure, sits, call, response, query, applications, fail, ways, existing, overjoy, before, burned, gic, made, making, live, self, correction, possible, boosted, issues, issue, resolution, 000, research, workflows, trusted, teams, rely, monthly, downloads, 1bn, stars, 60k, copy, cli, wizard, uvx, set, credit, card, required, 10m, telemetry, records, month, continuously, insights, traces, protect, spend, ships, bring, written, ship, resources, press, kit, logomark, logo, skip, main, content,

Text of the page (random words):
pydantic the end to end ai engineering stack skip to main content download logo as svg download logomark as svg go to press kit logfire ai gateway pydantic ai products enterprise pricing resources contact us sign up sign in pydantic ai build agents you ll actually ship to production trace production agents turn failures into eval cases and improve what ships with pydantic logfire bring your agents written in python or typescript continuously improve your agents with insights from your traces and evals protect your data and control your spend with logfire ai gateway start free explore enterprise no credit card required 10m telemetry records free every month set up logfire with your coding agent uvx logfire cli wizard copy from the team behind pydantic validation 60k github stars 1bn monthly downloads trusted across the python ecosystem and by engineering teams building the systems people rely on 12 faster issue resolution across more than 50 000 ai research workflows see how boosted ai fixes issues 12 faster 150 faster trace queries making live evaluation and agent self correction possible see how gic made trace queries 150 faster 20 cost spike caught before it burned through the team s budget see how overjoy caught a 20 cost spike problem ai applications fail in production in ways existing tools can t see the answer can be wrong even when every service is healthy evals that ignore production are exercises in creative writing they test the cases you imagined while the real failure sits in a tool call model response api or database query ai tracing model output looks healthy missing context application monitoring the database timeout is elsewhere solution one trace across your agent model calls apis and database logfire keeps the full execution path together so you can debug what broke evaluate what worked and improve the agent from real production evidence agent model api database one request one source of truth outcome one trace gives every team a better decision move from a symptom to the evidence that explains it without stitching together separate tools or guessing from averages evaluation synthetic test sets miss the failures users actually hit turn production failures into eval cases and verify the fix engineering tool sprawl turns incidents into archaeology follow one request through every layer of the application cost retries and model choices quietly burn budget see cost in context and cut what is not earning its keep security sensitive data can cross more systems than expected inspect the path enforce policy and retain the evidence production evidence logfire brings evidence to ai engineering see the agent s decisions in the context of the systems they touch logfire connects model calls and tool use to the apis databases queues and infrastructure behind the result so evals and fixes start with what actually happened integrations logfire works with your entire stack built on opentelemetry with first party sdks for python javascript and typescript and rust trace agent frameworks application code databases and the infrastructure underneath them without rebuilding your application around an observability vendor agent frameworks follow agents and tool calls across verified opentelemetry paths without tying production visibility to one framework pydantic ai langgraph crewai autogen google adk claude agent sdk strands agents mastra see the support matrix explore agent observability python models ai libraries openai anthropic google genai litellm mcp web frameworks fastapi django flask starlette asgi wsgi databases postgresql psycopg sqlalchemy asyncpg pymongo redis mysql javascript typescript vercel ai sdk browser next js express node js deno cloudflare workers rust built on the tracing and opentelemetry ecosystem tracing opentelemetry log crate any language via opentelemetry any framework with opentelemetry instrumentation works our sdks are convenient wrappers not a lock in boundary go java net ruby php erlang elixir swift c infrastructure keep the services hosts clusters containers and cloud metrics underneath each request in the same investigation dashboards alerts elasticsearch mongodb redis postgresql kafka nginx and more compute hosts kubernetes docker aws lambda metrics cloud providers system metrics explore infrastructure monitoring you can tell that pydantic logfire was built by people who use it dennis griffin vp of engineering at sophos start for free book a demo our mission building the best developer experience on planet earth pydantic did not start as a business idea it started with a developer problem worth solving well the team behind the validation library now brings the same obsession with clear apis useful defaults and open standards to the whole ai engineering stack about pydantic work with us i started working on pydantic out of frustration that type hints did nothing at runtime and curiosity about whether they could validate data pydantic s growth means the maintainers behind it now get to build more products on the same principle the most powerful tools can still be easy to use samuel colvin creator of pydantic samuelcolvin latest articles see all articles 22 october 2026 run pydantic ai coding agents in e2b sandboxes pydantic ai 8 mins laís carvalho 16 october 2026 pydantic runs on runson engineering 4 mins bill easton 13 october 2026 the code is the graph pydantic monty 9 mins samuel colvin 8 october 2026 announcement pydantic v2 14 release pydantic validation 15 mins victorien plot 5 october 2026 would you deploy an agent you don t control pydantic ai 6 mins aditya vardhan product logfire ai gateway pydantic ai enterprise pricing use cases llm observability llm evals llm as a judge ai gateway controls agent governance app performance apm service monitoring infrastructure monitoring frontend observability service levels slos open source pydantic validation pydantic ai pydantic evals monty explore all oss community blog customer stories events newsletter support documentation github slack company about us careers contact security compare alternatives logfire vs braintrust logfire vs langfuse logfire vs langsmith logfire vs arize logfire vs datadog logfire vs sentry logfire vs grafana logfire vs signoz logfire vs honeycomb logfire vs elastic logfire vs dash0 sign up for newsletter updates across the pydantic stack email subscribe something went wrong please try again you re in thanks for subscribing 2026 pydantic services inc privacy terms cookies trust center