An underwater observation company. Minnesota & Hawaiʻi. Founded 2026.

A workforce
for the living water.

N° 01     First field deployments, freshwater. Boundary Waters & Lake Superior.
Continue
N° 02   The Premise

We believe the living water deserves continuous, transparent observation — and we are building the workforce to provide it.

— Observatory, founding statement, 2026
N° 03   Approach

The observatory model, applied to water.

A traditional observatory is not a telescope. It is a long-running institution that points telescopes at the sky, records what it sees with care, and publishes the record so that others can build on it. The instruments change. The practice of observation does not.

Underwater observation deserves the same posture, and has rarely been given it. The vehicles that descend into reefs, kelp forests, lakes, and littoral zones are increasingly capable. The institutional practice around them — the way the resulting facts are documented, signed, owned, and shared — has not kept pace. We are building both.

i.

The substrate is open.

Hardware moats are increasingly fragile. Component supply chains, fabrication techniques, and reference designs are converging across the industry, and the conservation customers we serve cannot afford lock-in regardless. Our vehicles are built on published designs, documented bills of materials, and open firmware. The artifact is not the moat.

ii.

The record is sovereign.

Every observation an Observatory vehicle produces is signed at the moment of capture — vehicle identity, pod identity, firmware hash, calibration state, timestamp. The signed record travels with the data forever. The customer owns it. Two decades from now, anyone holding the public key can verify that a 2027 recording is what the recording claims to be. Provenance is infrastructure, not metadata.

iii.

The labor is the product.

We do not sell vehicles. We sell deployments — reef-weeks, lake-seasons, harbor-quarters. A customer pays for continuous observation of a place, performed by a coordinated fleet operated by people who have done this before, with the data flowing into a canonical record that the customer owns and the public can verify. The vehicle is a means. The continuous record is the end.

These three commitments are load-bearing. They constrain the company more than they free it, which is the point. An observatory that is not open, not sovereign, and not labor-shaped is a vendor with a marketing department.

N° 04   The Observatory

What we are hearing.

Status Pre-deployment · First field run Pending — Q3 2026, freshwater · Next update Within 30 days

The Observatory is the public face of every deployment we run. When a fleet is in the water, this page shows what the fleet is doing in close-to-real time — vehicle positions, recent observations, signed records you can verify in your browser. When no fleet is deployed, it shows the most recent observations and the full archive of every deployment the company has ever produced.

Below is the structural form of an Observatory record, populated with synthetic sample data so that the shape is legible before the first deployment fills it.

Listener pod · LP-001 Sample / Synthetic

Lake sturgeon drumming, candidate detection

A low-frequency rumble detected at 14m depth, matched against a candidate sturgeon vocalization profile with moderate confidence. In a live record, this card would link to the verified audio clip and the full detection trace.

CANDIDATE · 0.72 2 kHz 1 kHz 0 Hz
Capture2026-09-14 03:42:18 UTC
Location48.0539° N, 91.3621° W (offset 50m)
Depth14.2 m
VehicleFS-002 · firmware 0.3.1-rc4
Signing keyed25519 · 7c4a…3b91
Verify→ verify in browser
Deployment digest Synthetic

Six-day freshwater pilot

The form a deployment summary takes once the fleet has surfaced and the records have been signed, verified, and archived. The same record persists for the life of the company.

Site(reserved)
Dates(reserved)
Fleet1 tender · 3 free-swimmers
Pod hours
Observations
Verified
Public— / —
Every deployment leaves a permanent public record at this address. The Observatory accumulates rather than refreshes.
The Observatory is operational in form before it is operational in fact. We are publishing the structure of the record before the first deployment in order to make the company's commitment to provenance and openness available for inspection from day one. The first real record will replace this sample no later than the end of Q3 2026.
Enter the field simulator

Pilot a free-swimmer through six real water bodies under live conditions. Turn on live data to seed each site from active NOAA and USGS stations.

N° 05   Special Emphasis

The wrecks of Lake Superior.

The inland sea is the finest freshwater preserve of shipwrecks on Earth. Its cold water keeps hulls, timber, and painted lettering intact for a century and more. Every wreck is a documented fact — a known name, a known date, a documentation history that travels with it. That makes Lake Superior the truest possible subject for a provenance-first observatory, and our founding emphasis.

History

Identity, launch, final voyage, the storm that took her. Each record is source-stamped to the archive it came from.

Dive & condition

Depth, orientation, state of preservation — and the live conditions over the site right now, drawn from the nearest station.

Conservation

Protection status, monitoring needs, threats — the wreck as a fragile, non-renewable cultural resource.

Explore the wreck record Live Wisconsin data · curated Minnesota & Michigan
N° 06   Fleet

Mixed fleets, open hardware,
cognition at the edge.

A real workforce has roles. Our deployments combine tethered tenders — anchored, high-power, comms-rich — with battery-powered free-swimmers that range, listen, and dock periodically to the tender for charge and data offload.

Every vehicle runs the same three-tier compute architecture: a deterministic reflex layer that handles thrusters, leak detection, and emergency surface; a coordination layer that runs the mission state machine and acoustic communications; and a cognition layer — Raspberry Pi 5 for acoustic work, Jetson-class for vision — that handles the bioacoustic matching, anomaly detection, and the natural-language interface to the operator.

The vehicles are built on Blue Robotics components and our own pressure-housing designs. The CAD, BOMs, and firmware are published. Anyone can build one. Almost nobody can run a fleet of them well — and the work of running them well is what we sell.

Deployment topology · synthetic
SURFACE 10 m 20 m 30 m SURFACE TENDER STARLINK · GPS · COMMS T-001 · ANCHOR VISION POD · JETSON FS-001 LISTENER · PI 5 FS-002 LISTENER · PI 5 FS-003 SAMPLER SEABED

Sensor pods at launch

i.

Listener

Four-channel hydrophone array, Pi 5 cognition, closed-loop signal matching. Direct inheritance from our ORCA bioacoustic platform.

ii.

Surveyor

Stereo camera with onboard lighting and Jetson-class classification. Reef, kelp, lake-bottom mapping with verified provenance per frame.

iii.

Environmental

CTD, dissolved oxygen, pH, turbidity, fluorometry. The unglamorous baseline every customer asks for first.

iv.

Sampler

Small manipulator with water and sediment collection. The slow, expensive one. Scheduled for v2.

N° 07   News

The record begins with these notes.

Observatory is in its earliest phase. There is nothing here yet because there is nothing yet to report. The first entries will document our freshwater bench work, the construction of the first listener pod, and the path to the first field deployment.

— First posts expected, summer 2026
N° 08   About

An institution under construction.

Observatory was founded in 2026 to build the practice of underwater observation that the moment requires: continuous, transparent, sovereign, and open at the substrate. We are headquartered in Minneapolis with active institutional ties in Hawaiʻi.

The company emerges from a longer body of work on federated data infrastructure, provenance-forward systems architecture, and bioacoustic signal analysis — work that has already produced research nodes in Hawaiian botany, archaeological survey, and museum publication. Observatory is the marine and freshwater node of that federation. The three commitments on the Approach page are not aspirational; they are the way the rest of the federation already operates.

We are starting in freshwater because freshwater is where we live. Minnesota's ten thousand lakes, the Boundary Waters, and Lake Superior are not a detour from a saltwater story — they are the first chapter of an observatory whose subject is the living water in every form it takes.

Founded2026
BasedMinneapolis, Minnesota
TiesHonolulu, Hawaiʻi
StagePre-product · Pre-deployment
StructureTo be determined
FundingFounder · Grant pipeline in development
DomainFreshwater, then coastal saltwater < 100 m
First fieldQ3 2026 · Minnesota
Contacthello@[domain].org
N° 09   Careers

Not hiring yet.

Observatory will hire its first team in the second half of 2026, after the first freshwater deployment has produced a record that can stand on its own. The roles below are the shape of that team. If any of them describe work you have spent years preparing for, write to us now.

Underwater systems engineer

Pressure housings, thruster integration, leak detection, the unglamorous reliability work that decides whether a fleet survives a season.
Future — H2 2026

Bioacoustic researcher

Closed-loop signal matching, species-specific detection profiles, the soundscape literature. Comfortable working across freshwater and marine taxa.
Future — H2 2026

Field operations lead

Owns the practice of deployment — boats, batteries, calibration, the discipline of producing a clean record under field conditions.
Future — H2 2026

Provenance & data infrastructure

The signing-at-the-vehicle stack, the canonical schema, the public record. Has thought seriously about what a fact is and what makes it durable.
Future — H2 2026