Labric: structured data for scientific labs

What it does

Labric is data infrastructure for scientific research labs. It automatically captures instrument data, preserves its link to samples, protocols and experimental conditions, and turns scattered files into structured datasets that researchers can query and analyze with AI. The YC profile ↗ describes the data-layer thesis; the homepage ↗ shows instrument ingestion, team-wide visibility and natural-language questions across experiments.

The fit is a materials, chemistry, biology or other research organization where experimental context is scattered across instruments, Excel, notebooks and ad hoc databases. The buyer may be a lab leader or R&D platform owner, but the daily user is a scientist. That distinction matters: a beautiful dashboard is not enough if the instrument mapping, sample identity and reproducibility trail are unreliable.

Why I’d look closer

Labric publishes a rare set of concrete implementation claims. It says the platform supports 300+ instruments, streams and processes new data in a median eight seconds, and delivers 20x faster experimental cycle times versus traditional solutions. Those are company-reported figures, not an independent benchmark. The more important design detail is that each measurement is linked to its sample, protocol and experimental conditions, so context should survive team changes and cross-experiment analysis.

The docs ↗ expose a real developer surface: files, datasets, jobs, instruments, experiments and dashboards; API access can query data, manage files and trigger Python jobs. The docs describe scoped API permissions and isolated job sandboxes with ephemeral credentials. That is useful evidence for an evaluator, while still leaving the buyer to verify tenancy, retention and instrument-specific permissions.

The founder background is close to the pain. The About page ↗ describes Caitlin Hogan building a research-data platform at a materials-science startup for four years, with materials science and computer-science training, and Connor Hogan coming from database products and AI in Search at Google. Their first-hand lab-infrastructure experience is more relevant here than generic AI credentials.

What I’d ask

Which instruments are supported without custom engineering, and how are calibration, schema changes and missing metadata handled? I’d run one experiment end to end, export the raw and structured records, and test whether a new scientist can reproduce the analysis. The public site says “deploy in one week” and offers enterprise and academic paths, but the sources checked did not expose a standard price card.

My editorial take

Labric is a good fit for labs where data wrangling is the hidden bottleneck to AI-assisted discovery. The product deserves a closer look because it treats context and reproducibility as first-class objects. I’d validate one instrument family and one real protocol before believing the broad speed claim.

Quick facts

Sources checked

Checked 2026-09-19.

Cohort context

Labric is listed in Spring 2025. In our 2026-09-18 directory snapshot, 97 of 143 listed companies in that cohort have YC’s primary industry label B2B (67.8%). This is a current-directory comparison, not an original intake count or a performance ranking. Nine-cohort dataset.

Public website snapshot

Observed 2026-09-19 16:15 UTC in raw homepage HTML. This records visible metadata and advertised links, not agent execution or product quality.

Public observations · Collection method. Missing links here do not establish that a capability or file is absent elsewhere.