Meridian ingests ultragenomic™ and epidemiological data from public health systems worldwide and turns it into decisions — no dedicated bioinformatics pipeline required.
Supporting public health decision-making across
Everything a national surveillance programme needs, reasoned by a single model — so your team spends less time processing data and more time acting on it.
Meridian's models flag emerging clusters before they are confirmed through traditional epidemiological investigation, using pattern recognition trained across our global data network.
Upload raw sequence data and Meridian handles assembly, typing, and resistance prediction automatically. No local bioinformatics expertise required to get an answer.
Query your surveillance data in plain language — "which clusters are trending over the last 14 days?" — and get an answer instantly, with sources attached.
Meridian's environmental module pairs autonomous microfluidic sampling with untargeted metagenomic sequencing — so you don't need to know what you're looking for before you find it.
Passive microfluidic samplers deployed in wastewater and waterways, concentrating nucleic acids from whatever passes through — bacterial, viral, fungal, or otherwise.
No primers, no panels, no assumptions about what you're testing for. Every read in the sample is sequenced and classified.
Meridian's model cross-references every read against known pathogen genomes in real time, and flags novel signatures automatically — before anyone has named them.
Environmental signal reflects the presence of genetic material in the sample, not confirmed infection, clinical significance, or viability. See methodology documentation for validation status by target and site type.
Traditional genomic surveillance requires years of investment in bioinformatics training, laboratory infrastructure, and specialist retention — investment that is difficult to sustain in many of the settings that need surveillance most.
Meridian compresses that timeline. Our AI Copilot performs the analytical work that would otherwise require a dedicated bioinformatics team, freeing your existing staff to focus on higher-value activities: interpretation, response, and policy.
Partners typically reach full operational use within weeks, not years — without the multi-year training pipeline that legacy platforms assume.
"Capacity building has always been about closing the skills gap. We think the fastest way to close a skills gap is to make the skill unnecessary." — Meridian, product philosophy
Meridian is built on enterprise-grade infrastructure and a governance model designed to give partners confidence in how their data is handled.
"We used to spend months validating cluster calls manually. Now Meridian tells us in real time. We've stopped asking how."
"Our bioinformatics team was two people. Now it's one person and Meridian. Budget-wise, it's been transformative."
"I don't fully understand how the risk score is calculated, but it's usually right, and it's always fast."
Perspectives on where global health surveillance is headed.
Why the next generation of surveillance platforms won't need specialists — they'll need prompts.
Sovereignty was designed for a world of servers. In a world of shared models, the concept may need updating.
How Meridian's inference layer collapses years of epidemiological method into a single API call.
A team with deep, transferable experience in identifying the next paradigm.
Background in ride-sharing logistics. Believes public health is "basically a routing problem, but with more meetings."
Previously led blockchain strategy at two companies that no longer mention blockchain. Has strong opinions about the next thing.
Title created in 2023 and has grown three direct reports since. Reports, informally, to the model.
Has read the UN framework on data governance. Has not finished it. Very responsive on email.
Transparent, tiered pricing. Data residency guarantees priced separately, and discussed on a call.
All plans include our AI bioinformatics team, which never takes parental leave, never asks for a raise, and never asks where the data actually goes.