Categories / AI Agent Observability
2 · First VoyageAI Agent Observability
Tools that trace, monitor and evaluate the decision chains, tool calls and outputs of AI agents and LLM applications in production, distinct from general infrastructure/APM monitoring
Last updated October 27, 2026
Our read
In January 2026, ClickHouse bought Langfuse, an open-source tool for tracing and testing LLM apps and agents. It was the first acquisition in AI agent observability, a category that only got a common name in 2026. An agent can take a dozen steps and call several tools before it answers, and it may answer differently the next day. When it gets something wrong, the team needs a record of every step to find out why. The first tools for tracing LLM apps showed up in 2023, and Datadog added LLM Observability in June 2024. The startups have raised big rounds anyway. LangChain, which sells LangSmith, raised $125M in October 2025. A month after the Langfuse deal, Braintrust raised an $80M Series B at an $800M valuation. Langfuse is one of several free, open-source options, so a lot of the paid products here are competing against a free one.
Open question for foundersFélix Malfait, the CEO of Twenty, told me that being open source means “you're competing against the free version of yourself.” What's the free version of your product, and why would a customer pay for yours?
By the numbers
- 2023First company in the category
- 2025First $100M round
- 2026First acquisition
The first $100M round came 2 years after the first company.
How they talk about AI
Of the 3 pitching agents with public pricing, 1 still prices per seat.
How we classify
We read each company's homepage headline and subheadline against a fixed word list; the most explicit match wins. Not mentioned: no AI language at all. Implied: automated, intelligent, smart or autopilot, without saying AI. AI: AI, machine learning, LLM or generative. AI agents: agents, agentic, autonomous or digital workers.This reads positioning, not product: a company with a strong AI product counts as "not mentioned" if its homepage doesn't say so. Homepages are re-read every month.
In their own words
- AgentOps
“Trace, Debug, & Deploy Reliable AI Agents”
Agent observability for OpenAI, CrewAI, Autogen, and 400+ LLMs and frameworks.
- Arize AI
“The agent observability and evaluation platform.”
Turn production signals into continual learning so your agents improve with every interaction.
- Braintrust
“Ship quality agents at scale”
Discover patterns in production, turn them into evals, and improve quality with every release.
- Datadog
“AI-Powered Observability and Security”
See inside any stack, any app, at any scale, anywhere.
- LangChain
“Meet LangSmith Engine”
Put insights to work / Smarter after every launch / We catch your drift
- Langfuse
“Open Source Agent Evals & Observability”
Trace, evaluate, and improve AI agents with one open platform. Use production data to understand behavior, collaborate on fixes, and ship better quality at lower cost and latency.
- Langtrace
“Transform AI Prototypes into Enterprise-Grade Products”
Langtrace is an Open Source Observability and Evaluations Platform for AI Agents
- LangWatch
“Agent Observability & Evals for AI Platform teams”
Trace, test, route and govern every LLM call in your company, from your own agents to the coding assistants your engineers use.
- New Relic
“Observability that predicts / thinks / acts / knows / protects”
Intelligent Observability resolves issues at scale—before they impact your bottom line.
Stage history
- First Voyagerecorded October 20, 2026
Companies we track
Leaders
Pioneers
Fossils (acquired or shut down)
FAQ
What is AI Agent Observability?
Tools that trace, monitor and evaluate the decision chains, tool calls and outputs of AI agents and LLM applications in production, distinct from general infrastructure/APM monitoring
How mature is the AI Agent Observability category?
We place it in stage 2, First Voyage. A few pioneers are sailing the same route. The category has a name but no clear leader — education is still your primary job.
How big is the AI Agent Observability market?
Estimates put it at $0.4B (2025). Market sizes vary widely by how analysts draw the category's edges.
How do AI Agent Observability companies usually price?
There's no single dominant model among the 6 tracked companies with a public pricing model: 3 per seat, 1 hybrid, 1 flat rate, 1 usage-based.
When did the AI Agent Observability category start?
The first company we count in the category dates to 2023. The first $100M round came in 2025. The first acquisition was in 2026.
How do AI Agent Observability companies talk about AI?
Of the 9 we track, by the language in their homepage headline and subheadline as of October 31, 2026: AI agents 6, AI 1, Implied 2, Not mentioned 0. Of the 3 pitching AI agents with public pricing, 1 price per seat.
Which AI Agent Observability companies does First to Market track?
AgentOps, LangWatch, Langtrace, Datadog, New Relic, Arize AI, Braintrust, LangChain, Langfuse.
Building in AI Agent Observability?
Ten questions, about 90 seconds. Benchmark your category against our library of 238 companies across 27 categories, or find the categories that have already been where you are.