Capture specification
Agree the device or head-mounted camera, point of view, resolution, frame rate, audio rule, session length, task list, environment, metadata and prohibited content. Feasibility is assessed for the named cities and settings.
AI data services · Latin America
Egocentric video collection in Latin America
For AI and robotics teams that need first-person recordings of real tasks, Savia can recruit, onboard and manage contributors in Latin America project by project. Capture hardware, activities, locations, duration, camera position, metadata and acceptance thresholds are negotiated from your protocol. There is no open funded collection promised by this page.
Scope the work
People and places
Define contributor criteria, household composition and where recording may occur. Each adult household member signs consent under the applicable local process, children remain out of frame, and private spaces that cannot be recorded are documented during onboarding.
Pilot and acceptance
A proposed project starts with supervised test sessions. The parties agree how blur, occlusion, camera angle, incomplete tasks, interruptions and other rejection reasons affect acceptance and payment before scale-up.
What you receive
- Accepted media transferred to the storage and folder structure agreed in the project.
- Session metadata and acceptance status against the negotiated capture specification.
- Consent record linked to each contributor, with deletion handling defined in the agreement.
- Project report covering recorded and accepted hours, rejection reasons and reported incidents.
Quality and controls
- Onboarding and support are delivered in the contributor’s local language.
- Trainers review a daily sample against the buyer’s acceptance criteria and feed rejection reasons back to contributors.
- A contributor advances from a test session only after that session meets the agreed review threshold.
- Recruitment targets, trainer ratios, timing and price are proposed only after the protocol and locations are reviewed.
Savia Carga · Colombia
Inspect the method in a published release
Savia Carga publishes its tasks, answer key, derivations, rejected item, sources and limitations. It shows how Savia separates source evidence, deterministic computation, human confirmation and model inference.
Open Savia Carga ↗AI data services · Latin America
Questions about egocentric video collection
What collection specifications can you support?
Specifications are negotiated for each project. Send the device, camera position, video and audio settings, activities, locations, metadata, target hours and acceptance criteria; Savia will respond with what appears feasible.
How is household consent handled?
The operating model requires consent from the contributor and every adult in the household, keeps children out of frame, and records spaces that may not be filmed. The written process is adapted to the law and project location.
Is a collection project open now?
No open funded project is promised here. Savia recruits for a buyer’s funded, signed scope and tells prospective contributors the rate and permitted use before recording starts.
Who owns the recordings?
Ownership, licence, permitted uses, retention and deletion are set in the project agreement. Delivery goes to the storage arrangement specified by the buyer.
Discuss a pilot
Send the task, market, source material, rubric and expected volume. We will define what can be staffed and verified, then price the scope in writing before work begins.
Discuss a pilotAI data services
Services for AI and data teams
Spanish AI evaluation for Latin American markets
Regional Spanish AI evaluation for es-CO, es-MX, es-AR and other scoped variants: rubric scoring, preference ranking and red-teaming.
Explore the serviceExpert annotation for healthcare, law and finance
Healthcare, legal and finance annotation in Latin America with project-specific practitioner verification, source rights and item-level provenance.
Explore the serviceRegional benchmarks built from verifiable sources
Latin American AI benchmarks built from authorized sources, explicit derivations and regional task definitions, with model-inferred content clearly labeled.
Explore the service