Felipe Gazolla.
PT
← Projects
Real product experienceNov 2024 — present

Fito Agrícola.

Experience on a real product: AgroTech, mobile and geospatial data.

  • AgroTech
  • Mobile
  • SaaS

Farmers and consultants need to monitor properties, fields, crop management, inputs, weather and technical content — previously across fragmented flows and often tied to the computer.

A platform with a mobile app (Flutter), a web panel (Next.js) and an API (Laravel) that centralizes georeferenced properties and fields, crop management, reports, agronomic calculators, maps, weather and content — from the field to the office.

What I built.

  1. Podcasts feature (full-stack, end to end)

    I built the entire podcasts feature: Laravel backend (chapters, notes, status, search, likes and presigned URL upload to S3), web panel (player with chapters and play counter) and Flutter app (offline player, mini player, library and consultant area).

    • Laravel
    • Next.js
    • Flutter
    • AWS S3

  2. AI chapter extraction

    A pipeline that transcribes audio with Whisper and generates chapters and notes with GPT-4.1-mini, with transcription caching to reduce processing cost.

    • OpenAI
    • Whisper
    • GPT-4.1-mini
    • Laravel

  3. Field maps and geospatial data

    Drawing field polygons on the map, selecting coordinates and exporting/importing KML (boundary), over Google Maps and Leaflet, in a product that also integrates satellite imagery via Google Earth Engine.

    • Google Maps
    • Leaflet
    • KML/GeoJSON
    • MySQL Spatial

  4. Fungicide calculator

    Efficacy ratings and disease weights for corn and wheat (including diplodia), modifiers and availability per crop via API — agronomic logic applied across web, mobile and backend.

    • Laravel
    • Next.js
    • Flutter

  5. Crop management, reports and units

    Grouped load tickets and harvests, unified export per field, PDF reports (inventory, inputs, visits) and unit conversions (acre↔hectare and dose ha↔acre) respecting the property's unit of measure.

    • Laravel
    • DomPDF
    • Next.js
    • Flutter

  6. Weather forecast with spraying window

    Forecast charts and indication of the ideal spraying window in the app, helping the farmer plan applications.

    • Flutter
    • fl_chart

Technologies I worked with.

Mobile
FlutterDartRiverpodgoogle_maps_flutterfl_chartFreezed
Web
Next.jsReactTypeScriptSCSSChart.jsFramer MotionSWR
Backend
LaravelPHP 8MySQLJWTRedisAWS S3DomPDFExcel
Geospatial
Google Earth EngineGoogle MapsLeafletKML/GeoJSONMySQL Spatial
AI
OpenAIWhisperGPT-4.1-mini

Skills I developed.

  • Full-stack development (web, mobile and API) on a product in production
  • Mobile with Flutter and Riverpod
  • Backend in Laravel/PHP with MySQL and spatial data
  • Geospatial data: maps, polygons and KML
  • AI integration (Whisper and GPT) in a product
  • Agronomic domain: crop management, fungicides, units and harvests

How it connects.

  • Three applications on top of the same API: a Flutter (Riverpod) mobile app for the farmer in the field, a Next.js web panel for management and consulting, and a Laravel/PHP API with MySQL.
  • Geospatial field data (polygons) handled with MySQL Spatial and KML/GeoJSON; maps in Google Maps and Leaflet; satellite imagery via Google Earth Engine.
  • Media on S3 with presigned URL uploads, caching and queues in Redis, PDF generation with DomPDF and Excel exports.
  • AI pipeline for content: audio transcription with Whisper and chapter and note generation with GPT, with caching to reduce cost.

This work connects my background in Agronomy, my experience in agtech and my work as a full stack developer — a real product in production, with AI, geospatial data, mobile and APIs.