Home / Directory / AI SaaS tooling / Observability & monitoring / Laminar
Laminar
London company from Y Combinator's Summer 2024 batch that traces AI agents step by step, finds failures without preset rules, and lets developers rerun an agent from any step.
Engineering teams building long-running AI agents that want open-source tracing and debugging with low posted monthly prices.
Teams monitoring classic machine learning models, or enterprises that want a long-established vendor.
Verdict
Laminar is a London company that traces and debugs AI agents, showing each model call, tool call, and sub-agent in one readable view. It suits teams whose agents run for many minutes and fail in ways that are hard to reproduce. Browser Use and OpenHands run it in production.
Score Breakdown
How Laminar scores in the categories that matter to its buyers.
Pricing
Laminar publishes monthly plans with included data and per-gigabyte overage, plus a free tier, and quotes Enterprise.
| Plan | Price | What stands out |
|---|---|---|
| Free | $0 | 1 GB data, 7-day data access, 1 project, 1 seat |
| Starter | $30 / month | 3 GB data, then $2 per GB; 30-day data access; unlimited seats |
| Pro | $150 / month | 10 GB data, then $1.50 per GB; 6-month data access |
| Enterprise | Custom | On premises, custom limits, dedicated support |
The Field at a Glance
How Laminar compares with other Observability & monitoring vendors we reviewed, by Overall Score and relative typical engagement cost.
Laminar scores 7.9 in Observability & monitoring, under Fiddler AI (8.1), level with Langfuse (7.9) and Raindrop (7.9), and ahead of Arize AI (7.7) and Evidently AI (6.8). Relative cost lands lower for this subcategory.
Use-case matrix
| Use case | Fit | Notes |
|---|---|---|
| Agent tracing | Strong | Captures model calls, tool calls, sub-agents, and costs. |
| Rerun from any step | Strong | Restarts an agent with its prior context. |
| Open source | Strong | Written in Rust and self-hostable. |
| Low entry price | Strong | Free tier and $30 Starter plan. |
| Classic ML monitoring | Weak | Laminar traces agents and does not monitor model drift. |
| Vendor size | Weak | A young team founded in 2024. |
Review Excerpts
Excerpts from company testimonials, customer stories and product pages for Laminar. Paid reviews and pay-for-play sites were excluded.
Laminar says it stores only the unique content in agent runs, which it says makes storage 20 times cheaper.
“We run millions of agent sessions in our cloud, and when something goes wrong, Laminar's trace view is the first place we look.”
Laminar's Free plan keeps data for only 7 days and allows 1 project and 1 seat, and its Signals failure detection is not included.
Company Facts
- Founded
- 2024 (Y Combinator)
- Headquarters
- London, United Kingdom (Y Combinator)
- Accelerator
- Y Combinator, Summer 2024 batch (Y Combinator)
- Seed funding
- $3M led by Atlantic.vc, March 2026 (Tech.eu)
- Founders
- Robert Kim (CEO) and Dinmukhamed Mailibay (CTO) (Tech.eu)
- License
- Open source, written in Rust (Y Combinator)
- Website
- laminar.sh
Key Leadership
Robert Kim
Co-founder and CEO
Methodology
This page is an independent evaluation of Laminar for buyers comparing options in observability & monitoring. AI Industry Reviews accepts no sponsorships, advertising, or pay-for-placement fees. Laminar did not pay for this review.
What we scored
The headline number is an Overall Score on a 0-10 scale. Eight criteria sit under it in two groups.
- Buyer outcomes (for observability & monitoring)
- Agent trace debugging
- Open source and self-hosting
- Cost for small teams
- Vendor maturity
- Company & commercial
- Innovation & product leadership
- Project management & communication
- Pricing
- Contract fairness
Pricing measures whether the price looks fair for the value delivered, including packaging and renewal friction that show up in real buying cycles.
Score Composition
| Input | Weight | What it covers |
|---|---|---|
| Reviews | 40% | A proprietary read of what practitioners say about likes, complaints, and day-to-day use, including public review sites, forums, and private chat rooms. Paid reviews and pay-for-play sites are out of scope. |
| Product | 35% | Hands-on look at screens and workflows. |
| Pricing | 15% | Whether the price looks fair for what you get. |
| Docs & training | 10% | Docs, tutorials, and training material. |
How we balanced the evidence
The Overall Score is the simple average of the eight criteria. Recommendation language follows that score and the use-case matrix above.
Scope
Laminar is graded here as observability & monitoring. Criteria scores can move as more review volume and product checks are added.