Treat My Fish
Growers send photos, location, and water readings from a mobile app. Vizier drafts a diagnosis. Fisheries experts review the case in a web dashboard.
- Grower app
- Expert workflow
- Applied AI
Growers send fish-health cases to fisheries experts.
Fish growers use a mobile app to send photos, location, water readings, and feed data. Fish-health experts at SKUAST-Kashmir's College of Fisheries review each case in a web dashboard, send advice, and record follow-up.
The dashboard groups cases by species, district, and syndrome. Experts can see possible disease clusters. Reviewed cases can be exported for research and model testing.
- Role
- Product + full-stack + AI
- Interfaces
- Grower app and expert dashboard
- Institution
- SKUAST-K College of Fisheries
- Domain
- Coldwater fisheries: trout, carp
Across the screens.
Still frames from the interface demonstration. The examples use synthetic data.
Desktop demonstration
Water telemetry, a synthetic AI draft, and an expert handoff. A demonstration of the interface, not live model validation.
Mobile demonstration
Estimate daily feed from demonstration stock and fish-weight inputs.
How it works.
Vizier uses fish-pathology references for each draft diagnosis.
Selected fish-pathology references are stored in a searchable library.
The system retrieves passages that match each case.
Vizier uses the grower's photos, water readings, and those passages to prepare a draft.
An expert reviews the draft before sending the advisory.
Vizier prepares a draft using case data and fish-pathology references. An ichthyopathologist confirms the case before the advisory is sent to the grower.
Inside the system.
The flows, sources and review points behind the interface.
From pond-side to advisory
An ichthyopathologist confirms the draft. The reviewed case supports both the grower advisory and a view of cases by species, district and syndrome.
- GrowerPhotos and water readings
- Case
- VizierDraft diagnosis
- IchthyopathologistConfirmation
Parallel paths:
- AdvisoryTreatment for the grower
- SurveillanceSpecies, district and syndrome
How the reference library supports a case
Fish-pathology literature is prepared as a searchable reference library. Vizier combines matching references with the new case to prepare a draft.
Parallel paths that join at the next step:
- Fish-pathology literature
- Ingest and chunk
- Vectorize knowledge basefish_health domain
- New case
- VizierUses fish-health references
- Draft diagnosisReady for expert review
Context travels with the case.
Collect more than a photograph.
Water readings, location and feed data accompany the grower’s images. Fisheries experts receive that context in the web dashboard alongside the AI draft and its references.
Connect review with the wider picture.
Cases are grouped by species, district and syndrome. Experts can inspect possible clusters while keeping the individual case, advisory and follow-up together.
Keep reviewed cases available for research.
Expert-reviewed cases can be exported for research and model testing. The draft-to-review boundary remains explicit throughout the workflow.
Built around
the job.
Fish health system.
- Mobile app for growers and web dashboard for experts
- Cases include images, water readings, location, and feed data
- Searchable library of fish-pathology references
- Draft diagnosis based on case data and fish-health references
- Cases grouped by species, district and syndrome
- Expert-reviewed cases can be exported
Technology
React 18 · TypeScript · Vite · Capacitor · Cloudflare Workers (Hono) · D1 · R2 · Vectorize · Azure AI · Firebase Auth
From the product.
System overview drawn from the existing project record. The illustrated workflow is not an application screenshot.
Watch the interface film
Water telemetry, a synthetic AI draft, and an expert handoff. A demonstration of the interface, not live model validation.
Current local review build · Synthetic demonstration · 2 Oct 2026. This recording uses synthetic demonstration data; it does not establish current release status or real-world performance.
Watch the mobile recording
