Fun
The rest of my life: side projects, motorcycles, Japan, coffee and music. Open anything.
Mixtapes
One tape per project, and one mix. Pick one up.
Fonema
A pronunciation coach I owned from the first question to a working app
A daily pronunciation practice for Portuguese speakers learning American English. I started it for my wife and me, to work on our accents, and took it from research to a working app.
Overview
Brazilian learners often know the words and still get asked to repeat them.
The gap is a few sounds that Portuguese does not have, or makes differently: an extra vowel after a final consonant, the R at the start of a word read as an H, the L at the end turning into a W.
So how do you show which sound was off, and the movement that fixes it?
And how do you say so honestly when the engine cannot tell? Fonema is a web app that helps a learner say a word right on their own, in a few minutes a day, and shows whether they are improving. It teaches, not only grades: hear the model, say the word, see exactly what to fix, try again, and see real progress.
My role
There was no brief, no teacher on the team and no learner data, so I started from evidence.
I watched 20 videos by teachers who work with Brazilian students and counted what they name. The three most named problems:
- The R at the start of a word said like an H, so "rug" becomes "hug" (11 of 20 videos).
- The L at the end of a word turning into a W (9 of 20).
- An extra vowel after a final consonant, so "work" becomes "workee" (8 of 20).
The counts shaped which sounds the course teaches. From there I designed a learning model of seven rungs for every sound, levels built from those rungs, a placement test, and exams that give no feedback and compare the learner with their own first attempt.
I also read what already exists (BoldVoice, ELSA Speak, Speak and Duolingo) from their public pages, and chose a narrower ground: one language pair built around Brazilian errors, honest feedback, and no stored voice.
Decisions
Key product decisions
I began with no curriculum, no feedback model and no agreed rules for the game layer, so I made each question a written decision before building against it. I kept a log of 80 numbered decisions, each with a reason and a status, and set the scope by what the product refuses: no ads, no leaderboard, no hearts and no random rewards.
Decision 1
Follow the word the learner is saying
Instead of one mouth-view example per lesson, which in testing showed the same example for every word, I decided to show the mouth view per word: the word's sounds in order, the full movement for the sound being checked, opening by itself after a miss.
Decision 2
Say "can't tell" instead of guessing
Instead of guessing a score when the recording is too quiet, too short, clipped or noisy, I decided to say which one it was, and when it does judge, check one sound at a time and show the match percentage as an estimate next to the verdict.
A wrong "you got it wrong" costs more trust than a missing answer, even if a learner sometimes gets "move a little closer to the microphone".
Decision 3
The model speaks first
Instead of letting a learner say a word with no model, and then practise their own guess, I decided to have the voice say the word before the record button unlocks, and again before every retry. Exams do the opposite: no model and no feedback, so they measure.
Decision 4
One rule for the game layer, with four tests
Instead of leaderboards, hearts and random rewards, I decided to let a game element earn its place only if it shows the learner something true about their speaking, or gives them a choice.
Every idea had to pass four tests: does it report something real, can the learner ignore it, does anything get taken away or make them feel guilty for stopping, and does it stay quiet in an exam. A streak with up to two freezes passed, as did personal records and awards in tiers that only go up.
Engineering x design
I directed the build with Claude Code.
I decided what to make and why. It wrote and tested code under my direction, and the decision log, the research notes and the tests were the gates every change had to pass. The scoring engine builds on open source: OpenPronounce by Jean-François Lépine (MIT licence), the wav2vec2 phone model and the Kokoro voices (Apache-2.0). The product, the course design, the game layer, the research and the interface are mine.
Outcome
A working app with 626 words, 80 decisions and 935 tests.
A placement, lessons with a mouth view for every word, level exams, streaks with freezes, personal records and awards, a Second Wind round, sign-in by invitation, and a privacy design that keeps no recordings. It is in closed alpha, with 626 course words and their sounds, 80 logged decisions with their reasons, and 935 automated tests.
What is next
A closed alpha with Brazilian learners and English teachers.
Next comes a review of the course by a phonetician, and one thing the game layer cannot tell me yet: whether practice time and the clear rate move together. Then other first languages, starting with Spanish, each with its own error list and its own course instead of a translation.
Outrider
Smartwatch navigation that keeps a rider's hands on the handlebars
A self-initiated project: navigation that is safe and glanceable for urban motorcyclists, on the wrist instead of a phone.
Overview
Stopping to look at a phone is unsafe, and so is a phone mount.
I started Outrider to address a gap in accessible, safe navigation tools for motorcyclists, especially where stopping to consult a phone, relying on traditional motorcycle panels or using a phone support is impractical or unsafe.
So the navigation moved to the wrist.
The goal was a user-friendly, efficient and safe navigation tool with real-time GPS and trip information directly in a smartwatch interface: easily accessible, highly visible and minimally distracting, for riders in high-risk or complex urban environments.
Research
Understanding riders guided every stage.
I started with user interviews at the Interlagos Motorcycle Festival in São Paulo, one of the largest motorcycle festivals in Latin America, to uncover pain points and preferences. I also interviewed motorcycle dealers and brand ambassadors, and looked at gaps in smartphone navigation apps and built-in dashboards, and how people actually use them on a motorcycle. Being a motorcyclist and a smartwatch user myself also shaped a product that unites both worlds.
- Visibility and accessibility. A significant concern was having to take their eyes off the road to read navigation instructions.
- Safety. Riders were wary of phone mounts, because of theft risk and distraction, but a smartwatch app usually needs two hands, which is more dangerous even at low speed.
- Simplicity. People preferred a straightforward system: only essential information, high contrast and minimal detail, with minimal interaction.
Ideation and prototyping
A classic look with modern function.
I took inspiration from vintage motorcycle restoration and brat-style culture to find an authentic design direction. Sketching and wireframing explored layouts and interaction models for the watch. Then came a card sorting experiment and a heuristic evaluation of existing panel systems and curve-by-curve tripper systems, and an interactive Figma prototype simulating Outrider on a wearable, using the Apple Watch as the starting point.
- A simple, glanceable UI. Large, easy-to-read directions and a minimal design for quick reading, without losing information.
- "Twist of the Wrist" gesture controls. The pronated turn of the wrist, which most smartwatches recognize, changes screens between curve-by-curve navigation and the mini map, so riders keep their hands on the handlebars.
Try the gesture: drag the grip
Curve-by-curve navigation. Always on, focused on the next action the rider needs.
Mini map. The second screen, one twist of the wrist away.
Drag the grip left or right with a mouse or a finger, or use the arrow keys. In Outrider, a turn of the wrist switches the screen, so hands stay on the handlebars.
Final design
High contrast, always on, and quiet.
- High-contrast, minimalist interface. Legible in different lighting conditions, with less detail to save battery and avoid distraction.
- Turn-by-turn navigation. Always on and clearly displayed, focused on the next action the rider needs.
- Alerts and warnings. Vibrations for upcoming hazards or changing weather, so riders look at the screen less.
- Compatibility and a neutral profile. Outrider relies on existing apps and structures, using the watch's own design system and adapting it only where needed.
Testing
I tested it at the Interlagos Motorcycle Festival.
I took Outrider back to the festival where the research started, and showed the prototype to riders there, listening to what they made of it.
Photos: Festival Interlagos. Visit the site
Feedback
Riders responded well.
Outrider received very positive feedback from the motorcycle community. Riders said the system integration would make navigating dangerous or unfamiliar areas safer and more confident, without stopping or using their phones. In discussions, people especially appreciated the intuitive design and the practicality of the gesture controls.
It already fits casual riders and daily commuters. Delivery riders and moto-taxis are the next step, once their phone apps support smartwatches.
Garage
Garage
Where the world goes quiet.
Motorcycles are a big part of my life, and I like to tinker with them.
Racing drivers describe a moment when everything goes quiet. The noise drops away and the next corner is all that is left. A motorcycle gets me there on an ordinary road, because nothing sits between me and the weather, the surface or the sound of the engine.
Ewan McGregor and Charley Boorman took that feeling from London to New York in Long Way Round, east across Europe and Asia. My rides are shorter, on a Triumph Bonneville T100.
Looking after a bike is part of the ride. I listen for what changed, check the small things and learn what normal sounds like. It is the same habit I bring to design QA.
Vintage restoration and brat-style culture pull me in most, and they shaped the look of Outrider. That project started on the road: stopping to check a phone is unsafe, so I moved the navigation to the wrist.
Zen
Zen
Zazen and Zen Buddhism.
Japanese culture and Zen Buddhism are a big interest of mine.
Zazen
Zazen is seated meditation, the practice at the center of Zen. You sit upright and still, and pay attention to the breath. That is most of it.
- Seat. A cushion on the floor, or the front third of a chair. The full lotus is not required.
- Back and hands. Spine upright, shoulders loose, hands resting in your lap.
- Breath. Count each exhale from one to ten. When you lose count, start again at one.
- Mind. Thoughts will come. Notice them, return to the breath, and begin again.
Zen meditation
There is more than one way to sit. Beginners usually count the breath. Sōtō Zen is known for shikantaza, “just sitting”: no counting and no object, only sitting. Rinzai Zen is known for koans, short stories and questions meant to be sat with rather than solved by reasoning. Between sittings, kinhin is slow walking meditation.
Sit for a few minutes
05:00
A bell marks the start and the end.
Miragem
An umbrella brand for small coffee producers from Minas Gerais
Miragem is a coffee club and an umbrella brand identity for small producers from the interior of Minas Gerais, Brazil, a project initiated by Guia do Café São Paulo. I did the art direction, the identity design and the illustration.
Coffee and me
Coffee is part of where I come from.
Brazil grows roughly a third of the world's coffee, and Minas Gerais, home of Miragem's producers, grows about half of the Brazilian crop. Coffee is also where the story of Japanese Brazilians begins: the first ship of immigrants, the Kasato Maru, reached Santos in 1908 with about 780 workers for coffee farms in the state of São Paulo, many of them from Okinawa. I am a fourth-generation Japanese Brazilian, and Okinawan culture is part of where I come from.
I have gone to the São Paulo Coffee Festival every year. It celebrates the coffee scene of the biggest city in Latin America: tastings from the city's cafés and micro-roasters, workshops, live latte art and a roastery on site.
Photos: São Paulo Coffee Festival. Visit the site
Overview
Small producers made excellent coffee, but could not stand out in the city.
Miragem was a project initiated by Guia do Café São Paulo, as a way of empowering small coffee producers from Minas Gerais, Brazil. The producers faced significant challenges entering urban markets, primarily because they lacked a cohesive brand identity. Known for their high-quality, artisanal coffee, they struggled to distinguish themselves in a market dominated by well-established brands.
The challenge
One umbrella brand that city consumers connect with, and that stays true to its roots.
The primary challenge was to develop an umbrella brand identity that resonated with urban consumers' values while staying true to the roots of the Minas Gerais coffee culture. The brand needed to convey the quality, tradition and community spirit of the small producers in a way that was both authentic and appealing to a city and online audience.
Process
We started by listening to the producers.
After gathering insights and inspirations from our visits, we organized workshops with some of the producers. The aim was to uncover their needs, values and visions for the future together. The workshops revealed the deep connection between the land, the people and their coffee. The stories of generational farming, community support and the artisanal approach to coffee production became the cornerstone of our work.
The brand
A coffee club named Miragem, the Portuguese word for mirage.
The brand captures the essence of Minas Gerais coffee culture. Its elements combine to represent the coffee plant and the warmth of the community, chosen to reflect the product's organic and artisanal qualities. It focuses on sustainability and adapts easily to different materials and packaging, bridging the gap between local artisans and urban consumers.
Photoshoot
The identity on tape, a cup, a bag and a tin.
