Circus, Ingredient Portioning
Redesigned a tablet app that improved ingredient tracking accuracy during preparation.
Apr 1, 2025
Company
Circus Group
Company
Circus Group
Role
Senior Product Designer
Role
Senior Product Designer
Service
Led end-to-end Design
Service
Led end-to-end Design
Duration
4 weeks
Duration
4 weeks
Business Context
Business Context
In our robotics kitchen company, ingredient preparation wasn’t automated, leading to inefficiencies, safety risks, and costly errors from inaccurate silo placement and missing ingredient traceability, a challenge that would only grow with scaling.
The goal: create a system that guides kitchen staff step-by-step as they fill silos, ensuring every ingredient is placed correctly, while logging time and location to keep stock accurate, traceable, and error-free.
In our robotics kitchen company, ingredient preparation wasn’t automated, leading to inefficiencies, safety risks, and costly errors from inaccurate silo placement and missing ingredient traceability, a challenge that would only grow with scaling.
The goal: create a system that guides kitchen staff step-by-step as they fill silos, ensuring every ingredient is placed correctly, while logging time and location to keep stock accurate, traceable, and error-free.
Problem
Problem
Before the app, operators relied on Excel sheets and memory to manage silo prep. The robot couldn’t detect what was inside each silo, it only recognized position. When silos were misplaced or mislabeled, recipes failed, meals were wasted, and allergen risks increased. The workflow was slow, inconsistent, and not scalable for a commercial kitchen.
Before the app, operators relied on Excel sheets and memory to manage silo prep. The robot couldn’t detect what was inside each silo, it only recognized position. When silos were misplaced or mislabeled, recipes failed, meals were wasted, and allergen risks increased. The workflow was slow, inconsistent, and not scalable for a commercial kitchen.
Impact of Research
Impact of Research
By shadowing operators during full prep sessions, we uncovered where the workflow broke down. Operators had to juggle Excel, shelves, scales, and silos at once, creating constant context switching and confusion. The Excel interface itself was hard to read, with small text and no real-time validation, making errors easy and invisible until too late. Because the menu rarely changed, operators relied heavily on memory instead of structured guidance. Slot positions existed only in Excel, with no visual feedback during prep, making it easy to mix up silos or misplace ingredients. These observations exposed the need for a guided, error-proof flow that matched real operator behavior and provided clear feedback at every step.
By shadowing operators during full prep sessions, we uncovered where the workflow broke down. Operators had to juggle Excel, shelves, scales, and silos at once, creating constant context switching and confusion. The Excel interface itself was hard to read, with small text and no real-time validation, making errors easy and invisible until too late. Because the menu rarely changed, operators relied heavily on memory instead of structured guidance. Slot positions existed only in Excel, with no visual feedback during prep, making it easy to mix up silos or misplace ingredients. These observations exposed the need for a guided, error-proof flow that matched real operator behavior and provided clear feedback at every step.
Approach & Rationale
Approach & Rationale
The project started with alignment sessions involving product, operations, and culinary teams. Together, we defined the essential prep steps: select an ingredient, set the expiration date, enter the weight, print a label, and place the silo in the correct position. We chose to build a minimal, focused MVP to validate the flow early, rather than over-designing the interface before confirming the basics worked in practice.
The project started with alignment sessions involving product, operations, and culinary teams. Together, we defined the essential prep steps: select an ingredient, set the expiration date, enter the weight, print a label, and place the silo in the correct position. We chose to build a minimal, focused MVP to validate the flow early, rather than over-designing the interface before confirming the basics worked in practice.
Execution
Execution
The app was designed for clarity and precision. Each screen guided operators through a single step with large touch targets, clear feedback, and visual progress indicators. A label preview helped verify details before printing, and slot numbers ensured correct placement on the robot. The system logged every action ingredient, weight, expiry, and time, creating a full traceability loop. During usability testing, all participants completed the flow successfully, surfacing small but meaningful improvements such as clearer tare messaging and faster access to reprint labels.
The app was designed for clarity and precision. Each screen guided operators through a single step with large touch targets, clear feedback, and visual progress indicators. A label preview helped verify details before printing, and slot numbers ensured correct placement on the robot. The system logged every action ingredient, weight, expiry, and time, creating a full traceability loop. During usability testing, all participants completed the flow successfully, surfacing small but meaningful improvements such as clearer tare messaging and faster access to reprint labels.
Outcomes
Outcomes
The Silo Preparation App replaced unreliable Excel workflows with a structured, traceable system. Operators can now prepare silos faster and with complete confidence in ingredient accuracy. The new flow eliminated placement errors, reduced prep time, and provided reliable data for inventory tracking and operations analysis.
In parallel, I collaborated with the Operations team to design the prepping station setup, testing ergonomics and hardware layout to ensure the tablet, scale, and printer fit seamlessly into the workflow. We added position numbers to the trolley to improve clarity and prevent mix-ups. Together with the Ops manager, we co-created a smart cart loading system that guides prep based on slot positions, ensuring each silo is placed exactly where the robot expects it.
We also worked with the Culinary Team and VP of Product to co-design the physical prepping station, a simple setup optimized for today’s manual process but built to scale toward automation. A key improvement was making the tablet height adjustable to improve operator comfort during long prep sessions.
The Silo Preparation App replaced unreliable Excel workflows with a structured, traceable system. Operators can now prepare silos faster and with complete confidence in ingredient accuracy. The new flow eliminated placement errors, reduced prep time, and provided reliable data for inventory tracking and operations analysis.
In parallel, I collaborated with the Operations team to design the prepping station setup, testing ergonomics and hardware layout to ensure the tablet, scale, and printer fit seamlessly into the workflow. We added position numbers to the trolley to improve clarity and prevent mix-ups. Together with the Ops manager, we co-created a smart cart loading system that guides prep based on slot positions, ensuring each silo is placed exactly where the robot expects it.
We also worked with the Culinary Team and VP of Product to co-design the physical prepping station, a simple setup optimized for today’s manual process but built to scale toward automation. A key improvement was making the tablet height adjustable to improve operator comfort during long prep sessions.
Learnings
Learnings
Designing for hybrid environments means thinking beyond the screen. Small details like label visibility, scale placement, and tablet height had a major impact on usability. Testing early with real users proved that even simple flows require iteration. The next step is to integrate smart scales to automatically record weights—pushing the process closer to full automation.
Designing for hybrid environments means thinking beyond the screen. Small details like label visibility, scale placement, and tablet height had a major impact on usability. Testing early with real users proved that even simple flows require iteration. The next step is to integrate smart scales to automatically record weights—pushing the process closer to full automation.
