Descartes Labs

Descartes Labs was a Santa Fe, New Mexico geospatial intelligence company founded in 2014 as a spin-out from Los Alamos National Laboratory. The company built the Descartes Labs Platform, a cloud-native geospatial data refinery and analytics environment combining a petabyte-scale satellite imagery archive (Landsat, Sentinel-1/2, MODIS, NAIP, PlanetScope, SkySat, commercial radar, NEXRAD weather radar, plus client-uploaded data) with a Python client library, JupyterHub-based workbench, and distributed compute for training and running deep learning and remote-sensing models at scale. Customers in agriculture, mining, energy, defense, insurance, and the U.S. government used the platform to build production geospatial machine learning pipelines covering crop forecasting, mineral exploration, infrastructure monitoring, methane detection, wildfire response, and ESG reporting. The Platform exposed a Catalog (imagery, bands, products, blobs, events), a Compute service (containerised functions, jobs, schedules), a Vector service (tabular and geospatial features), Dynamic Compute (lazy raster algebra and tiling), Auth, and a `descarteslabs` CLI. In October 2024, EarthDaily Analytics — backed by Antarctica Capital — acquired Descartes Labs and the Descartes Labs Government, Inc. subsidiary, folding the team, customers, and platform into the EarthDaily Constellation programme. The product was rebranded EarthOne in 2025; the `descarteslabs` Python package has been formally discontinued in favour of `earthdaily-earthone` (v5.x), the github.com/descarteslabs organisation has been renamed to `dlarchives`, www.descarteslabs.com is now a parked domain, and the docs.descarteslabs.com developer portal has been retired in favour of EarthDaily-hosted EarthOne documentation. This catalog entry preserves the historical Descartes Labs Platform surface as an archive — see the EarthDaily / EarthOne profile for the active product.

6 APIs 15 Features
GeospatialGeospatial IntelligenceEarth ObservationSatellite ImageryRemote SensingRasterVectorGISMachine LearningGeospatial AnalyticsAgricultureMiningEnergyDefenseClimateAcquiredEarthOneEarthDailyDiscontinued

APIs

Descartes Labs Platform (Archived)

The historical Descartes Labs Platform — a managed geospatial data refinery and analytics environment exposing imagery catalog, raster access, vector tables, compute functions, ...

Descartes Labs Catalog API

The Catalog organised all imagery and derived data on the Platform. Products group Bands which group Images; Storage Blobs hold arbitrary file artefacts; Events emit notificatio...

Descartes Labs Compute API

The Compute service ran user-supplied Python code as containerised Functions against the imagery archive at scale. Users defined a `Function` (CPUs, memory, environment, Docker ...

Descartes Labs Vector API

The Vector service hosted tabular and geospatial feature data as Tables of typed columns with `uuid` identifiers and ipyleaflet visualisation. Supported property filtering (incl...

Descartes Labs Dynamic Compute API

Dynamic Compute was the Platform's lazy map-computation engine for interactive raster algebra and tile rendering in notebooks. Expressions over imagery products (band math, mosa...

Descartes Labs Auth API

The Auth module handled token-based authentication against app.descarteslabs.com — OAuth login flow, refresh tokens, and the user namespace claim that served as a global identif...

Features

Cloud-native geospatial data refinery built on a petabyte-scale satellite imagery archive
Native ingest and time-aligned access to Landsat, Sentinel-1, Sentinel-2, MODIS, NAIP, PlanetScope, SkySat, commercial radar, and NEXRAD
Catalog service organising Products, Bands, Images, Storage Blobs, Events, and EventSchedules
Event-driven processing — NewImage / NewStorage / NewVector / compute-function-completed subscriptions delivered to SQS or Compute Functions
Compute service for containerised Python Functions, Jobs, and bulk `Function.map` submissions over the imagery archive
Vector service for tabular and geospatial feature tables with property filtering, `ilike` wildcards, and ipyleaflet visualisation
Dynamic Compute engine for lazy raster algebra and on-demand XYZ tile rendering in notebooks
JupyterHub-based workbench on app.descarteslabs.com for in-browser notebook authoring
Python client `descarteslabs` (PyPI) with auth, catalog, compute, config, core, geo, and vector subpackages
`descarteslabs` CLI for managing Products, Bands, Blobs, and sharing from the command line
Sharing model with owners / writers / readers and AuthCatalogObject permission helpers
DL-COVID-19 mobility dataset and Contrastive Sensor Fusion research releases
Enterprise customers across agriculture, mining, energy, defense, insurance, and U.S. government
Descartes Labs Government, Inc. subsidiary for U.S. federal workloads
Discontinued in 2025 and superseded by EarthDaily EarthOne (`earthdaily-earthone` 5.x) following the October 2024 acquisition

Resources

🔗
Website
Website
🔗
Documentation
Documentation
💬
Support
Support
🔗
Login
Login
👥
GitHubOrganization
GitHubOrganization
📦
SDKs
SDKs
📦
SDKs
SDKs
📦
SDKs
SDKs
💻
CodeExamples
CodeExamples
💻
CodeExamples
CodeExamples
💻
CodeExamples
CodeExamples
💻
CodeExamples
CodeExamples
🔗
Dataset
Dataset
🔗
Research
Research
🔗
PythonPackage
PythonPackage
🔗
PythonPackage
PythonPackage
🔗
Successor
Successor
🔗
Successor
Successor
🔗
Acquisition
Acquisition
🔗
Acquisition
Acquisition

Sources

apis.yml Raw ↑
aid: descartes-labs
name: Descartes Labs
description: Descartes Labs was a Santa Fe, New Mexico geospatial intelligence company founded in 2014 as a spin-out from
  Los Alamos National Laboratory. The company built the Descartes Labs Platform, a cloud-native geospatial data refinery and
  analytics environment combining a petabyte-scale satellite imagery archive (Landsat, Sentinel-1/2, MODIS, NAIP, PlanetScope,
  SkySat, commercial radar, NEXRAD weather radar, plus client-uploaded data) with a Python client library, JupyterHub-based
  workbench, and distributed compute for training and running deep learning and remote-sensing models at scale. Customers
  in agriculture, mining, energy, defense, insurance, and the U.S. government used the platform to build production geospatial
  machine learning pipelines covering crop forecasting, mineral exploration, infrastructure monitoring, methane detection,
  wildfire response, and ESG reporting. The Platform exposed a Catalog (imagery, bands, products, blobs, events), a Compute
  service (containerised functions, jobs, schedules), a Vector service (tabular and geospatial features), Dynamic Compute
  (lazy raster algebra and tiling), Auth, and a `descarteslabs` CLI. In October 2024, EarthDaily Analytics — backed by Antarctica
  Capital — acquired Descartes Labs and the Descartes Labs Government, Inc. subsidiary, folding the team, customers, and platform
  into the EarthDaily Constellation programme. The product was rebranded EarthOne in 2025; the `descarteslabs` Python package
  has been formally discontinued in favour of `earthdaily-earthone` (v5.x), the github.com/descarteslabs organisation has
  been renamed to `dlarchives`, www.descarteslabs.com is now a parked domain, and the docs.descarteslabs.com developer portal
  has been retired in favour of EarthDaily-hosted EarthOne documentation. This catalog entry preserves the historical Descartes
  Labs Platform surface as an archive — see the EarthDaily / EarthOne profile for the active product.
type: Index
position: Provider
access: 3rd-Party
image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg
tags:
- Geospatial
- Geospatial Intelligence
- Earth Observation
- Satellite Imagery
- Remote Sensing
- Raster
- Vector
- GIS
- Machine Learning
- Geospatial Analytics
- Agriculture
- Mining
- Energy
- Defense
- Climate
- Acquired
- EarthOne
- EarthDaily
- Discontinued
url: https://raw.githubusercontent.com/api-evangelist/descartes-labs/refs/heads/main/apis.yml
created: '2026-05-24'
modified: '2026-05-24'
specificationVersion: '0.20'
apis:
- aid: descartes-labs:descartes-labs-platform
  name: Descartes Labs Platform (Archived)
  description: The historical Descartes Labs Platform — a managed geospatial data refinery and analytics environment exposing
    imagery catalog, raster access, vector tables, compute functions, and authentication through a Python client library,
    REST endpoints under *.descarteslabs.com, and a `descarteslabs` CLI. The Platform was discontinued and rebranded as EarthDaily
    EarthOne in 2025. APIs and endpoints below are preserved for historical reference; no public OpenAPI specification was
    ever published.
  tags:
  - Geospatial
  - Platform
  - Archived
  humanURL: https://github.com/dlarchives/descarteslabs-python
  properties:
  - type: Documentation
    url: https://github.com/dlarchives/descarteslabs-python
  - type: SDKs
    url: https://github.com/dlarchives/descarteslabs-python
  - type: PythonPackage
    url: https://pypi.org/project/descarteslabs/
  - type: Successor
    url: https://github.com/earthdaily/earthone-python
- aid: descartes-labs:catalog
  name: Descartes Labs Catalog API
  description: The Catalog organised all imagery and derived data on the Platform. Products group Bands which group Images;
    Storage Blobs hold arbitrary file artefacts; Events emit notifications when new images, blobs, or scheduled triggers fire
    so that Compute Functions can be automatically invoked. Catalog supported sharing via owners / writers / readers, EventSubscriptions
    (SQS, ComputeFunctionCompleted, NewImage, NewStorage, NewVector), and EventSchedules for calendar-based runs. Backed by
    REST endpoints on *.descarteslabs.com and surfaced as `descarteslabs.catalog`.
  tags:
  - Catalog
  - Imagery
  - Products
  - Bands
  - Blobs
  - Events
  humanURL: https://github.com/dlarchives/descarteslabs-python/tree/master/descarteslabs/catalog
  properties:
  - type: SourceCode
    url: https://github.com/dlarchives/descarteslabs-python/tree/master/descarteslabs/catalog
  - type: Module
    url: https://github.com/dlarchives/descarteslabs-python/tree/master/descarteslabs/catalog
- aid: descartes-labs:compute
  name: Descartes Labs Compute API
  description: The Compute service ran user-supplied Python code as containerised Functions against the imagery archive at
    scale. Users defined a `Function` (CPUs, memory, environment, Docker image), submitted Jobs (often via `Function.map`
    over hundreds of thousands of tiles), and collected results. Compute integrated with Catalog Events to fire on new imagery
    and emitted `compute-function-completed` events on drain-to-zero. Surfaced as `descarteslabs.compute` (Function, Job,
    JobStatistics, JobBulkCreateResult, ComputeClient).
  tags:
  - Compute
  - Batch
  - Functions
  - Jobs
  humanURL: https://github.com/dlarchives/descarteslabs-python/tree/master/descarteslabs/compute
  properties:
  - type: SourceCode
    url: https://github.com/dlarchives/descarteslabs-python/tree/master/descarteslabs/compute
  - type: Module
    url: https://github.com/dlarchives/descarteslabs-python/tree/master/descarteslabs/compute
- aid: descartes-labs:vector
  name: Descartes Labs Vector API
  description: The Vector service hosted tabular and geospatial feature data as Tables of typed columns with `uuid` identifiers
    and ipyleaflet visualisation. Supported property filtering (including `ilike` wildcard matches), table options for column
    ordering, and tile-based rendering. Originally shipped as the standalone `descarteslabs-vector` PyPI package and absorbed
    into `descarteslabs.vector` in client 3.0.0.
  tags:
  - Vector
  - Tabular
  - Features
  - Visualization
  humanURL: https://github.com/dlarchives/descarteslabs-vector
  properties:
  - type: SourceCode
    url: https://github.com/dlarchives/descarteslabs-vector
  - type: Module
    url: https://github.com/dlarchives/descarteslabs-python/tree/master/descarteslabs/vector
- aid: descartes-labs:dynamic-compute
  name: Descartes Labs Dynamic Compute API
  description: Dynamic Compute was the Platform's lazy map-computation engine for interactive raster algebra and tile rendering
    in notebooks. Expressions over imagery products (band math, mosaicking, temporal compositing) were built lazily and evaluated
    on demand against the compute backend, with results streamed as XYZ map tiles into an ipyleaflet map. Distributed as the
    separate `descarteslabs-dynamic-compute` package.
  tags:
  - Dynamic Compute
  - Raster
  - Tiles
  - Map Algebra
  humanURL: https://github.com/dlarchives/descarteslabs-dynamic-compute
  properties:
  - type: SourceCode
    url: https://github.com/dlarchives/descarteslabs-dynamic-compute
  - type: PythonPackage
    url: https://pypi.org/project/descarteslabs-dynamic-compute/
- aid: descartes-labs:auth
  name: Descartes Labs Auth API
  description: The Auth module handled token-based authentication against app.descarteslabs.com — OAuth login flow, refresh
    tokens, and the user namespace claim that served as a global identifier for the authenticated user across Catalog, Compute,
    Vector, and Storage. Surfaced as `descarteslabs.auth` and consumed transparently by every other client module.
  tags:
  - Auth
  - Authentication
  - OAuth
  humanURL: https://github.com/dlarchives/descarteslabs-python/tree/master/descarteslabs/auth
  properties:
  - type: SourceCode
    url: https://github.com/dlarchives/descarteslabs-python/tree/master/descarteslabs/auth
  - type: Module
    url: https://github.com/dlarchives/descarteslabs-python/tree/master/descarteslabs/auth
common:
- type: Website
  url: https://www.descarteslabs.com
- type: Documentation
  url: https://docs.descarteslabs.com
- type: Support
  url: https://support.descarteslabs.com
- type: Login
  url: https://app.descarteslabs.com
- type: GitHubOrganization
  url: https://github.com/dlarchives
- type: SDKs
  name: Descartes Labs Python Client (descarteslabs)
  url: https://github.com/dlarchives/descarteslabs-python
- type: SDKs
  name: Descartes Labs Dynamic Compute
  url: https://github.com/dlarchives/descarteslabs-dynamic-compute
- type: SDKs
  name: Descartes Labs Vector (legacy standalone)
  url: https://github.com/dlarchives/descarteslabs-vector
- type: CodeExamples
  name: Tutorials
  url: https://github.com/dlarchives/tutorials
- type: CodeExamples
  name: Example notebooks
  url: https://github.com/dlarchives/example-notebooks
- type: CodeExamples
  name: Workflows examples
  url: https://github.com/dlarchives/workflows-examples
- type: CodeExamples
  name: Enterprise Accelerator onboarding notebooks
  url: https://github.com/dlarchives/descarteslabs-ea-notebooks
- type: Dataset
  name: DL-COVID-19 mobility dataset
  url: https://github.com/dlarchives/DL-COVID-19
- type: Research
  name: Contrastive Sensor Fusion for remote sensing
  url: https://github.com/dlarchives/contrastive_sensor_fusion
- type: PythonPackage
  url: https://pypi.org/project/descarteslabs/
- type: PythonPackage
  name: descarteslabs-dynamic-compute on PyPI
  url: https://pypi.org/project/descarteslabs-dynamic-compute/
- type: Successor
  name: EarthDaily EarthOne Python client (replaces descarteslabs)
  url: https://github.com/earthdaily/earthone-python
- type: Successor
  name: EarthDaily Analytics
  url: https://earthdaily.com
- type: Acquisition
  name: EarthDaily Analytics Announces Acquisition of Descartes Labs (Oct 2024)
  url: https://earthdaily.com/blog/descartes-labs-acquisition
- type: Acquisition
  name: PR Newswire — EarthDaily Analytics acquires Descartes Labs
  url: https://www.prnewswire.com/news-releases/earthdaily-analytics-announces-acquisition-of-descartes-labs-302276388.html
- type: Features
  data:
  - Cloud-native geospatial data refinery built on a petabyte-scale satellite imagery archive
  - Native ingest and time-aligned access to Landsat, Sentinel-1, Sentinel-2, MODIS, NAIP, PlanetScope, SkySat, commercial
    radar, and NEXRAD
  - Catalog service organising Products, Bands, Images, Storage Blobs, Events, and EventSchedules
  - Event-driven processing — NewImage / NewStorage / NewVector / compute-function-completed subscriptions delivered to SQS
    or Compute Functions
  - Compute service for containerised Python Functions, Jobs, and bulk `Function.map` submissions over the imagery archive
  - Vector service for tabular and geospatial feature tables with property filtering, `ilike` wildcards, and ipyleaflet visualisation
  - Dynamic Compute engine for lazy raster algebra and on-demand XYZ tile rendering in notebooks
  - JupyterHub-based workbench on app.descarteslabs.com for in-browser notebook authoring
  - Python client `descarteslabs` (PyPI) with auth, catalog, compute, config, core, geo, and vector subpackages
  - '`descarteslabs` CLI for managing Products, Bands, Blobs, and sharing from the command line'
  - Sharing model with owners / writers / readers and AuthCatalogObject permission helpers
  - DL-COVID-19 mobility dataset and Contrastive Sensor Fusion research releases
  - Enterprise customers across agriculture, mining, energy, defense, insurance, and U.S. government
  - Descartes Labs Government, Inc. subsidiary for U.S. federal workloads
  - Discontinued in 2025 and superseded by EarthDaily EarthOne (`earthdaily-earthone` 5.x) following the October 2024 acquisition
  sources:
  - https://github.com/dlarchives/descarteslabs-python
  - https://pypi.org/project/descarteslabs/
  - https://earthdaily.com/blog/descartes-labs-acquisition
  - https://www.prnewswire.com/news-releases/earthdaily-analytics-announces-acquisition-of-descartes-labs-302276388.html
  - https://github.com/earthdaily/earthone-python
  updated: '2026-05-24'
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com