February 4, 2025

Gain R2 and R2 Pro data-driven coffee roaster to access the Aillio Eco system
Gain R2 and R2 Pro data-driven coffee roaster to access the Aillio Eco system — 1
Gain R2 and R2 Pro data-driven coffee roaster to access the Aillio Eco system — 2

Overview

Key Features

  • Data Structure: Working with the Firmware team to design and implement a data stream for every component and sensor
  • Display and Command: Design how to control and display the roast chart of AiO, the data-driven coffee roaster, by cross-platform desktop apps, from UI (Vue.js) to send the command through the communication layer (Golang) 
  • Integration: Make the new roaster able to collect the roast profile, automatically sync data to the cloud, use the roasting recipes, and exist on the company dashboard website Roast.World
  • Took a flight to the head office in Taiwan with the team to set up the showcase machine at the showroom

Technologies Used

  • Vue.js (Typescript): For simplifying code style and real-time user interaction
  • Golang: use for a performance module, use as a C wrapper, and be the main package parser for this project
  • Electron: wrapper for all services to run as cross-platform desktop apps
  • Websocket: the communication method for internal service, and RestAPI for the sync
  • Firebase: as the main source of truth for the sync system
  • Vitest/Playwright: the project included unit tests, some user testing cases, and hardware integration tests
  • Figma: designing the UI by following the company handbook, and creating UI for prototypes

Challenges and Learnings

This plan comes from the management team's decision after market analysis. No know this before, and we had a tight deadline, just about 6 months for the entire company, and my team had only 3.5 months left when the testing board was handed to us. At that time, we were trying to make our desktop apps as ready as possible before the testing board came. We opened the urgent call, where we tried to develop software 24 hours/ 5days among our global team. And we make a clean and simple document to share the development/test/pending to pass this

Outcome

We're able to ship the first batch of R1 in Q1 2025, with full functionality. We have made R2 able to roast with our program RoasTime, able to save roast profiles, using recipes, have our advanced logging, analysis, and gain access to the Roast.World online community and roasting history dashboard