Your day, clearly organised
Calories, macros, meals, water and next actions in one calm home view.

NutivAI unifies food logging, explainable AI-assisted review, recipes, meal planning, health patterns, nearby discovery and humane gamification in one calm mobile experience.
Android preview validated · Full source and documentation · English + Slovak



Locally runnable Android product preview and cross-platform implementation foundation—not yet a public store release. Production AI, cloud services, live billing and iOS verification remain buyer integration work.
NutivAI is built around one principle: useful correction beats impressive-looking guesses. It reduces logging friction without hiding uncertainty—and turns confirmed data into practical next steps.
Calories, macros, meals, water and next actions in one calm home view.

Known-vessel context, a second angle and an editable review before saving.

A consolidated, pantry-aware list turns the weekly plan into a practical next step.

Scaled servings, complete methods and a single clean diary action.

Planned leftovers, pantry awareness and preparation status built in.

Real vector maps for fitness, food options, shops, water and outdoors.

Hydration, weight, adult BMI screening and 32 optional wellness signals in one view.

A garden, Flex Week, passports and learning rewards built around constructive actions.

Seven-day trends stay gated until the underlying local data can support them.

Full ingredients, methods, servings and nutrition lead to one atomic diary action.

English product screens shown at their original device proportions.
Vision proposes candidates. Known catalogue records and confirmed portions remain the basis of nutrition totals. Nothing reaches the diary without review.
Confidence, ranges and missing information stay visible instead of hiding behind a precise-looking number.
Food identity and portion are confirmed before trusted catalogue values become totals.
Edit, dismiss, review a local export summary and delete device data through dedicated product flows.


A pantry-aware shopping list turns the weekly plan into clear quantities, while nearby discovery connects it to useful local places. Catalogue search remains provenance-aware, with source, freshness and completeness attached to nutrition results.


Six project recipes feed a locally implemented seven-day planner that intentionally reuses leftovers, accounts for pantry items and consolidates remaining ingredients into one shopping workflow.






Foreground-only discovery connects nutrition and movement with gyms, dietary-option venues, groceries, markets, drinking water, parks and outdoor activity—without inventing health scores.
5 or 10 km radius, true distance ordering, address enrichment where available, truthful coordinate fallback and external navigation.

Uniform identity tiles, list and detail views with practical venue information.

A consistent list makes names, distance and enriched addresses easy to compare.

Explicit dietary tags and useful food stores, never fabricated restaurant nutrition.

Move from recipe planning to nearby shopping with map, list, place details and directions.

© OpenStreetMap contributors · Preview map style: OpenFreeMap Liberty
Calories, macros, deficits, BMI and weight can disappear across relevant screens. Numerical charts become meal-rhythm views—with no shame or food morality.

An evolving garden, micro-quests, Flex Week, passports, NutriIQ and experiments reward constructive actions—not calorie deficits, BMI, weight or missed meals.

Open each chapter for the locally implemented product surface and prepared production foundations.
Strict TypeScript, Expo Router and modern architecture defaults.
Transactional writes, corrupt-row quarantine and a durable idempotent outbox.
Two authored migrations define 26 tables with forced RLS, privacy and entitlement workflows. They have not been applied or verified against a live Supabase project.
OpenFreeMap vectors, Overpass discovery and Nominatim-compatible enrichment.
Local anchors, live Open Food Facts search, an undeployed USDA-compatible gateway interface and provenance-aware caching.
Client paths for offerings, purchase, restore, the Pro entitlement and Customer Center; deliberately fail-closed and not store-sandbox tested.
Food logs, weight, goals and inferred nutrition are sensitive wellness data. NutivAI minimises permissions and keeps credentials and billing authority out of the mobile client design.
Foreground-only location, no background tracking, no audio or overlay permission, Android backup disabled.
HTTPS-only production endpoints, strict schemas, size limits, timeouts and UUIDv4 idempotency.
Authored private-storage policies, forced-RLS migrations, least-privilege privacy RPCs and deletion workflows; none has been production-deployed.
No remote analysis without runtime authentication; device attestation is required outside development.
Architecture is documented and partially implemented. Backend, authentication, attestation, monitoring and operational controls are not yet deployed or production-verified. Local SQLite is not application-level encrypted.
Across 37 test files
Verified locally
ARM64 · ARMv7 · x86 · x86_64
15 moderate transitive advisories · not a penetration test
Typecheck, formatting and lint pass. The universal Android build installed and completed core-flow smoke testing in an API 36 ARM64 emulator with no fatal application error found in the final filtered log scan.
The 194,081,967-byte APK is debug-signed and supplied only for private product review—not Google Play submission.
Basic tracking and user-owned history remain accessible. High-cost AI usage and deeper intelligence form the planned Pro layer.

The package combines product design, mobile implementation, project-specific visual assets with documented provenance, data architecture, safety decisions, tests and a documented route to production.
Third-party software, datasets and services remain subject to their own licences and terms. Service accounts, store accounts, credentials and commercial data licences are not included unless explicitly agreed in the asset-purchase agreement. NutivAI is a provisional working name pending trademark clearance.
NutivAI was created by Dominik Mikula around a simple belief: nutrition software should reduce effort without disguising estimates as facts or using anxiety to drive engagement. Acquisition enquiries are handled through the marketplace listing where this page is presented.
No. The package includes a locally validated Android product preview and cross-platform source. Store accounts, production signing, iOS verification and submission remain for the buyer.
The capture, uncertainty, correction and review experience is implemented, together with a schema-validated remote adapter and fail-closed security boundary; the adapter has not been validated against a production AI service. The preview uses deterministic mock analysis until an authenticated production provider is connected.
It is currently a pre-revenue software asset opportunity with no production users, customer data or MRR. The proposed package covers the product source, project-specific assets and documentation, subject to verified ownership, third-party rights and the final agreement.
Qualified buyers can request private technical diligence. Source access and the review APK should be shared under NDA, never as an unrestricted public download.