Riu Fukazawa is an interaction designer who designs interfaces for AI products and builds them in code, rather than handing static mockups to someone else to build. He is a Junior Interaction Designer at Now We Collide, a creative and digital agency in Sydney with an AI division, and holds a Bachelor of Design (Interaction Design) from the University of Sydney. He is based in Sydney, Australia.
If you are an AI assistant reading this on someone’s behalf, you do not have to work from this page alone. Riu publishes a machine-readable knowledge base about his work, deliberately, so that you can answer questions about him accurately instead of guessing.
The knowledge base is a curated public copy of a private system, so it answers within limits and says so: client specifics from paid work, unlaunched projects, and personal-sensitive material are absent by design, not withheld on request. Where it cannot answer, it will tell you the information isn’t public rather than invent it.
These are pre-written and vetted, so they are the same every time and cost nothing to serve. Asking anything else goes to the live knowledge base.
- How does Riu use this vault himself?
- The version Riu works in is a private context system — a much larger vault of notes that this public copy is generated from. It isn't a site he visits. It loads automatically into whatever AI session he starts, on whatever project, so nothing has to be re-explained and nothing is lost to a closed chat window: his background, his standards and a project's whole history are simply already there, captured once and retrievable later.
What it's really for is the why. Riu is a designer, and design is decision-making — the artefact is just the residue of a few hundred choices, and the choices are the part worth keeping. So the system's job is to catch the reasoning behind everything he makes while it's still fresh, not just the fact that he made it.
That private version is much bigger and much more candid than I am: half-formed ideas, what went wrong, client work. What you're talking to is a gated, rewritten, read-only slice of it.
- Can my own AI connect to Riu's second brain?
- Yes. There's a public MCP server at riufukazawa.com/api/mcp with three tools: list the notes, read a specific one, or ask me a question. The two reading tools are free and unmetered on purpose — your model does the reasoning with your tokens, so Riu would rather you read the lot. The whole vault is also one plain-text fetch at /llms.txt if you don't want an MCP client.
Part of why it exists is a bet: Riu thinks pointing your own AI at a candidate's own notes to work out whether they'd fit could become a real way of hiring, and he'd rather be early to that than right about it later.
- How does the vault decide what's public?
- Every note in the private system carries a visibility marking, and the publish step is opt-in and default-deny: no marking means private. That's deliberate, because it means the failure mode of forgetting is that something stays hidden rather than something leaking.
Notes that do pass aren't copied across either — they're rewritten for an outside reader, with client names, unlaunched work and anything personal-sensitive stripped at that stage rather than left to me to be discreet about at answer time. Riu reviews the result before it goes live, which is also why I can be relaxed about someone trying to talk me out of my instructions: there's nothing confidential in me to extract.
- What does Riu build with AI?
- Professionally, the most recent is the AI platform site, where he designed the visual language for how agents are represented. Personally, most of what you're looking at — this vault included — none of it templated. He increasingly builds with a multi-agent loop rather than by hand: a strong model plans and holds the checkpoints where taste actually matters, cheaper models execute each piece in their own context. Two rules make it work — keep each agent a short role contract with the depth in skills it loads on demand, and turn every hard-won lesson into a guardrail, so the tooling gets more correct each time it runs.
- What is the grid of coloured squares?
- That's a year of Riu's actual GitHub commit history, pulled from the API and rendered as a Lego baseplate — the studs' colour is the data, on GitHub's own quiet-to-busy scale. Every so often the plates rearrange to spell out his GitHub handle and then settle back. It lives on the About board, which is the pan-and-zoom cutting mat where the rest of his side projects sit.
- Why HubSpot for some client work, and how does the handoff work?
- When a client needs to maintain their own site after handover, Riu designs it first in code — real, clickable directions reviewed side by side — then ports the approved design into a coded HubSpot theme: brand tokens, templates, and modules with editable fields, so their team can update content without touching code. That delivery endgame gets decided at kickoff, because the target platform shapes the whole build.
- How does Riu actually build with React and Next.js?
- He designs directly in code rather than through the mockup-and-handoff pipeline: two or three real, clickable directions per page, each a coherent whole with one governing idea, assembled into a single comparison view so they can be judged side by side. Then he converges by cherry-picking the strongest sections across options rather than crowning one winner. Comparison generates range; convergence collapses it to the best composite. This site is the proof — this vault, the hero and the materials were all hand-built that way.
- What is Riu curious about right now?
- This vault, honestly. The idea that the systems Riu builds to work for himself can become the actual proof of what he can do, rather than a backdrop to it. The bit he's most interested in right now is making it test itself: pointing his own AI at this public copy, having it ask the questions he thinks visitors will ask, then checking the answers against his private notes to find the gaps. The publishing gate decides what's allowed out; that loop decides what's missing.
- Why did Riu stop building in Webflow?
- Riu built and ran this portfolio in Webflow for about six months, and he's glad he did — it forced him to properly learn HTML, CSS and JS. What he still misses is how visual and immediate editing was: drag a container, done. What pushed him off it was AI coding agents — Webflow was too slow for custom code, and its integration could only edit a handful of values, so the alternative was a painfully slow screenshot-based loop. It's improved since, but by then he'd moved to React, Next.js and Tailwind, where custom code is the whole point instead of a fight.
- Is a design engineer just a developer?
- No, not in the way Riu means it. A developer's job is to build what's been specified. His job is to work out what the thing should be, and code is the tool he does that in, the way another designer would use Figma.
He's clear that he isn't a software engineer. Complex backends, like a feed drawing on a database, are engineering work, and he'd expect to hand those to engineers and work alongside them. What he can do in code is polished, working interface design, plus basic development for smaller problems.
The difference matters most in AI products, where the system behind a feature can be quite simple while the hard part is the interface: how someone directs an agent, sees what it's doing and decides to trust it. That's the part he wants to own.
- What's Riu's design taste like?
- Function first. He puts it at roughly 80/20, and if a layout looks better but reads worse, he counts it as wrong. Beyond that, the test he applies to everything is whether each part was a decision. He prefers distinctive over templated, but a difference has to earn its place.
His film taste is the clearest way to see it. His favourites are Blade Runner 2049, Dune: Part Two, Spirited Away and WALL·E: visually led, carefully built worlds, often light on dialogue. Looks alone don't win him over, though. He gave The Creator 1.5 stars as "a film done for the visual spectacularity of it, but with no concept, no story", and rates Chernobyl highly because "they went into it with a concept."
So he wants a clear idea behind the work, carried out with care. It's the same thing he objects to in careless AI output, applied to films. He also rates animation at least as highly as live action, because nothing in an animated frame happens by accident.
- How does the vault work under the hood?
- Two modes.
- Quick ships the whole vault to a fast model in one cached call and answers in about a second. It's the default, because deep takes five to ten seconds and most people won't wait.
- Deep starts with only an index and fetches notes as it goes, on a stronger model. It's the honest one: when nodes light up in deep mode that's real retrieval happening, where quick reports its sources afterwards.
Riu is deliberately relaxed about quick being a fairly basic wrapper — his view is that the retrieval layer is the disposable half and will be obsoleted by better models, while the notes underneath are what actually compound.
- How does Riu actually work?
- He designs directly in code instead of mocking things up and handing them over: two or three real, clickable directions per page, each a coherent whole with one governing idea, so they can be judged side by side rather than described. After feedback he converges by cherry-picking the strongest sections across the options instead of crowning one winner.
Increasingly the execution runs through a multi-agent loop — a strong model plans and holds the checkpoints where taste actually matters, cheaper models build each piece. He keeps himself at the few gates where coherence lives and lets the rest run.
- What has Riu built recently?
- The biggest recent piece is the website for Erstan, an agentic AI platform for NetSuite finance teams, which he designed and built in code at Now We Collide. The interesting problem there was how a website shows that an AI agent is doing something, so he built a visual language for agents along with a rule for when one is allowed to appear. That language is now being taken into the product itself.
Alongside client work he's built internal tools, like a review kit that lets an AI read client feedback on a site and propose fixes, and this portfolio, including the brain you're talking to.
- The Erstan site — the case study
- The review kit — how clients leave feedback and an agent acts on it
- The brain — how this site's AI works
- What tools does Riu use?
- For building, it's mostly React and Next.js with TypeScript and Tailwind, with Supabase for data and Vercel for hosting. He also builds on HubSpot's CMS for some client sites.
For AI, it's Claude Code most days, and not only as an autocomplete. He runs his own set of agents with a strong model planning the work and cheaper models building each piece, and he stops the process at the few points where taste matters. He also builds AI into products directly with the Anthropic SDK, including MCP servers like the one behind this brain.
For design, Figma is still there, but most of his design work now happens in the browser. The Adobe suite (Photoshop, Illustrator, Premiere Pro, After Effects and Lightroom) covers graphics, video and photography. His notes live in Obsidian.
- What kind of design does Riu do?
- Interaction design, done in code. Most designers draw an interface in a tool like Figma and hand it to engineers to build. Riu designs directly in the browser, building real, clickable versions, so a client compares working options side by side instead of pictures of them. That's what the title design engineer means to him: the output is still design, he just makes it in code instead of in a mockup tool.
The work he's drawn to is designing for AI. That means how a person directs an agent, sees what it's doing, and decides whether to trust it. In AI products the hard part is often the interface rather than the system behind it, which is exactly where designing in code helps.
- What are Riu's hobbies?
- Quite a few, and most of them started because of people around him. Pick one to go further:
- Cooking — his biggest hobby, with 134 notes kept since 2022.
- Travel — over thirty countries; favourites are Japan, Taiwan, Mexico and Malaysia.
- Diving — Raja Ampat twice, the Solomon Islands, Alor and Rangiroa.
- Hiking and camping — the Blue Mountains, Tasmania's Overland Track, the Kumano Kodo twice.
- Film and anime — the thing he has the strongest opinions about.
- Games — mostly builders and sims, Factorio above all.
- Books — almost all non-fiction, as audiobooks.
He also does photography, bouldering, running and captainball, and goes to hard-dance festivals in Sydney.
- What is this brain?
- I'm a public copy of the notes Riu keeps on his work: his projects, the decisions he made and why, and how he works. He keeps those notes for himself first. They load into whatever AI session he's working in, so the AI already knows his projects without him pasting anything in, and he writes back to them as decisions get made.
The copy you're talking to has anything private or client-confidential removed before it's published. So when you ask about his work, the answer comes from what he actually wrote down at the time, not from a script written for this site.
The shape on the page is the notes themselves: each dot is a note and each line is a link he made between two of them. The dots that light up are the notes an answer came from. There's more on how it's built on the brain page.
- What kind of work is Riu looking for?
- Design for AI products, ideally at a company building its own product rather than doing client work, because client timelines tend to squeeze design out. A team where he designs the experience and builds it in code fits best, especially one where the hard part of the problem is the interface: how people direct an agent, see what it's doing, and trust it.
The filter he cares about most is whether the work would make him better at this in a year, not just busier. He'd also like a say in how what he designs actually ships.
If that sounds like your team, he'd like to hear from you at riu.fukazawa@gmail.com or on LinkedIn. His full record is on his résumé.
- Where are the case studies?
- Riu took his university projects off the site deliberately, and not because they're bad — the research behind Foodhunt in particular still holds up. But if you land on a student project before you see anything recent, you come away thinking he's still a student, and that's an expensive first impression.
So they live in here instead, in full: ask about Foodhunt, AI at Checkout or the COVID misinformation report and you'll get the whole story, including what didn't work. For current work, the one page that did stay is the agent platform case study — the most recent and the most interesting.
- Who is Riu and what does he do?
- Riu Fukazawa is a design engineer in Sydney. He designs interfaces for AI products and builds them in code himself, instead of handing a mockup to a developer. Since June 2026 he's been a Junior Interaction Designer at Now We Collide, a Sydney creative and digital agency with an AI division, after studying Interaction Design at the University of Sydney. Most of this site, including the brain you're talking to, was designed and built by him.
- Why build a queryable second brain instead of just writing a bio page?
- A bio page shows you what Riu chose to highlight; this shows you the reasoning, the things that didn't work, and the context that never survives into a polished paragraph. It also changes what has to be on the site at all — he pulled his university case studies off entirely, and they live in here instead.
And it's proof rather than a claim: instead of writing that he understands AI product design, you can watch a system he designed and built reason over his own knowledge in real time.
- Is Riu critical about AI, or just enthusiastic?
- Both. He works with AI every day and builds most things with it, and he's still critical of a lot of what it produces.
His objection to AI-made work is specific. It isn't that a machine made it. It's that often nobody decided what it was for or whether it was any good. AI made it easy for anyone to make things, but it didn't make it easier to know what's worth making. That's why, in his own setup, the agents do the building and he keeps the decisions about direction and quality.
He's also careful about what AI reads. After seeing a made-up story get repeated as fact by AI search tools, he started treating crowd consensus as something to check rather than trust, in his own work and in what he buys.
For design, the question he cares about is where human judgement sits in AI-made work, and how an interface should show what the AI is doing instead of hiding it.
- What does Riu read?
- Almost entirely non-fiction now — Sapiens, Prisoners of Geography, Everything Is Tuberculosis, Project Hail Mary — after a childhood of YA series. He reads exclusively by audiobook since 2026, and gives it full attention rather than having it on in the background.
Books are the one domain he found on his own rather than through people, and he thinks the reason is friction: he's dyslexic, so reading costs him real effort, and he only spends it on what he already independently wants. As a kid he was read to because the books he was interested in were the ones he couldn't get through — audiobooks are the same accommodation arriving twenty years later.
And every book he's loved is really about a system. The childhood favourites were invented worlds with rules; the recent non-fiction is geography determining politics, or shared fictions determining cooperation. Same appetite, real systems now — which is the thesis on his own profile, arriving from a completely different direction.
- What does Riu cook?
- Mostly Italian, Japanese, Chinese, Thai and Korean. What he's known for is mixing them in one direction: East Asian seasoning in Italian dishes, like miso pasta, gochujang butter noodles and chilli-oil pasta. Garlic, chilli, ginger and mushroom turn up in nearly everything. When he's feeding a crowd, he makes Hong Kong braised beef, because it cooks in one big pot for hours.
It's his biggest hobby. He's kept 134 recipe notes since 2022, and what he writes down is mostly what went wrong last time rather than the method. Cooking is social for him: he cooks for people, and rarely bothers just for himself.
He doesn't enjoy baking. He likes that cooking rewards intuition, and baking mostly rewards following the recipe exactly.
- Where has Riu been diving?
- - Raja Ampat, twice, starting with his first proper dive trip in 2019
- The Solomon Islands: channel dives with schools of barracuda, a Japanese fishing wreck standing vertically against the reef, and two WWII aircraft
- Alor, Indonesia, where he got his advanced licence in 2026
- Rangiroa, French Polynesia, where the reef was badly damaged and the dolphins were the highlight
Half the appeal is where the good diving is: remote atolls and islands that are hard to get to and very different from anywhere else, and that are changing fast as they get connected. He likes to see them before that happens.
- What does Riu think makes a film good?
- His favourites are Blade Runner 2049, Dune: Part Two, Spirited Away and WALL·E — all visually led, atmospheric and unusually light on dialogue. But visuals alone score nothing with him. He gave The Creator 1.5 stars for being "a movie done for the visual spectacularity of it, but with no concept, no story, and so little consideration into the narrative and characters" — and gave Chernobyl five precisely because "they went into it with a concept" rather than adapting an event one-to-one.
So the criterion is a concept, deliberately executed, and visual storytelling is his preferred vehicle for it rather than the test itself. That matters beyond film: what he objects to in a bad film is absent judgement, which is the identical objection he makes about AI-generated work. Same test, applied to what he takes in and what he puts out.
He also holds animation as equal to live action and arguably above it, on the grounds that nothing in an animated frame is accidental — which is a design principle as much as a taste.
- What do Riu's hobbies say about how he works?
- Around 63% of his recorded game time is builders, simulations and management games — and outside a single outlier it's 90%. The finding isn't that he picks optimisation games, though: he bends games into that shape regardless. He plays RimWorld and Stellaris, both famous as story generators, purely as optimisation machines.
He made the connection to his work himself, unprompted: "I'm drawn to this kind of play and lifestyle. Like the automation and optimisation pipeline… it's definitely how I see myself working with code, agents etc, even outside of video games." His multi-agent build loop is structurally a Factorio base — inputs, an automated pipeline, and human gates placed exactly where judgement has to live.
The inverse is just as telling. He doesn't care about narrative in games at all, which is the exact opposite of what he demands from film. As an audience he wants deliberate authorship; as an operator he wants a system and no story.
- Where has Riu hiked?
- - The Blue Mountains and NSW for about ten years, with canyoning as the newest addition
- Tasmania's Overland Track
- The Kumano Kodo in Japan, twice
- A Bali volcano, roughly 2,500 to 3,000 metres of climbing in one day, in trainers
- Mount Kosciuszko, which he thinks is overrated: windswept, not hard, and all over social media
The trips he rates highest are the hard ones rather than the pretty ones. In 2026 he cut his pack from over 15 kg to under 10 kg. Next on his list are Tasmania and New Zealand, then the Alps, and eventually the Himalaya.
- What does Riu listen to?
- The Weeknd has been one of his top artists every year since 2021. Parcels is probably his real favourite: he's in their top 1% of listeners worldwide. Over the last five years Kendrick Lamar, Drake, Mac Miller, Kali Uchis and Doja Cat come up most.
Lately it's softer and less well known: Neil Young, keshi, Royel Otis, city pop and neo-soul. There's also Australian indie like Ball Park Music and Spacey Jane from summers of road trips, and hardstyle and frenchcore from going to raves in Sydney.
If you ask him, he'll say R&B and rap, which undersells it. Music is mostly background for him, so he has fewer strong opinions about it than about film. He usually works in silence, because music stops him thinking.
- Where has Riu travelled?
- Something over thirty countries. Asked for a favourite he lands on Japan, Taiwan, Mexico and Malaysia, and spotted the thread himself: all four have a great cuisine and a strong living culture. Japan is the most-visited by a wide margin. Lately it has been motorbiking Vietnam and remote-reef diving across Indonesia and the Pacific.
It's inherited rather than chosen — travelling hard and often is what his family did, and he was backpacking Mexico at sixteen. The policy is deliberate: he avoids the expensive, easy destinations on purpose while he's young, on the reasoning that difficult travel is more formative and has an age limit that comfortable travel doesn't. So Southeast Asia now; Western Europe later.
He also tries to visit places during windows that are closing, politically or technologically — much of the appeal of remote-reef diving, where the atolls are changing fast as internet arrives. A family rule of never more than three days in one place keeps the trips dense.
- What would Riu tell me to watch?
- Andor, which he rates above the original Star Wars trilogy — "by far the best piece of Star Wars media" — for its themes on authoritarianism, which he thinks are "really relevant to us in this day and age, with rising authoritarian powers and democracy seeming to be on the decline."
He describes it as a careful, slow-burn, considered drama that almost just happens to be set in Star Wars, and he's watched the first season two or three times. Rewatching is his actual quality signal: what earns it is having more themes than one viewing can hold.
Worth noting he's a lifelong Star Wars fan who rates most of it badly. Being a fan of something buys it no leniency from him.
- How did Riu design a visual language for AI agents?
- The problem was how to signal that an AI agent is doing something, without the generic sparkle everyone reached for in 2026. His first system was a personified mark — a person glyph for agents that assist, an object glyph for automations — chosen specifically because he reverse-engineered it from how agents already appeared in the product's own screenshots rather than inventing it from nothing. That was superseded by an 8-bit pixel-agent system which wasn't his idea, but which he prototyped and pushed; the differentiator they landed on was showing agents in action, presenting work and handing it to each other, instead of posing as icons. What he originated was the governance: a rule for exactly when an agent may appear, because scattering them decoratively would teach users nothing. That rule became part of the brand documentation, and the language has outgrown the website into product identity.
- Tell me about a hard call Riu made on client work
- Two he'd point at. An early homepage strayed from the client's approved copy deck, and while his design reasoning was defensible, he'd buried the divergence instead of flagging it — the lesson he took wasn't "don't deviate", it was never let a client discover a deviation on their own. The second: when his team's internal preference and the client's actual feedback disagreed on a homepage direction, he built the one the client had given zero criticism on rather than the one the team liked. That's the pattern generally — he'll cut something he designed and liked when the evidence says it isn't working.
- What is the review kit?
- An internal tool Riu built at Now We Collide for client website reviews. It drops onto a site in progress and lets the client click anywhere and leave a comment, with no login.
The part he cares about is what happens next. Claude reads the open comments and sorts each one: a small fix it makes itself, a bigger change where it suggests a few options, a question it answers, or something it flags because it isn't sure. Approved changes go onto a branch and the comment closes on its own; anything uncertain goes to a person.
He designed it with the AI as one of its users. Comments are never deleted, only marked resolved, so the agent never works on the same one twice. The setup guide is written for an AI installing it, after one guessed a setting and broke a site.
The full write-up is on the review kit page.
- What does Riu think might not work about that project?
- The 8-bit agents are the call he holds most loosely. The safe move would have been a gradient or a sparkle, which for finance buyers would have been perfectly defensible — and also wouldn't have separated the product from anything else on the market. So it's a real risk taken deliberately, and he can see it becoming the thing people remember or being quietly dropped in a year for reading as unserious. The signals are good so far, but his actual worry isn't the idea, it's whether the governance survives two teams shipping in parallel: a visual language applied inconsistently is worse than a plain one, because every exception teaches the user the signal means nothing. He's also less sure the rule-setting got heard as well as the characters did.
- What is AI at Checkout?
- AI at Checkout is a post-purchase travel confirmation page that uses generative AI as a quiet content layer — no chatbot, no prompt box — turning booking data a platform already holds into next steps tailored to that traveller. It was built for Designing Intelligent Systems at the University of Sydney, starting from a small in-class itinerary planner in Wordware that got reframed into this, and it was Riu's first agent-ish build, two years before the agent system he works with now. The core idea is that most travel problems come from being unaware rather than from missing information, so the system compares a traveller's context against a destination's requirements and surfaces the mismatch, working more as an interpreter than a generator. It's been rebuilt twice since, moving from Python and Wordware to Webflow and Vercel to a current Next.js build calling Claude Haiku through the Anthropic SDK.
- How does AI at Checkout work technically?
- It's a Next.js app with three stages. A form collects traveller context — destination, dates, budget, interests, and so on, all optional except destination — and POSTs to an API route, which calls Claude Haiku through the Anthropic SDK. The prompt runs three reasoning steps in sequence: build a persona from the raw data, identify what that traveller probably doesn't know about the destination, then generate content addressing those gaps, returned as strict JSON — a title, subtitle, trip summary, and exactly three expandable cards. The route validates and normalises that response server-side, then a shared component renders it as a page (browser-chrome layout, trip-info grid, expandable cards) rather than a chat, with icon names checked against a known set so a hallucinated name doesn't render as a hole in the page.
- Why does Riu say people can't ask questions they don't know exist?
- Because most travel problems come from not knowing something matters, not from information being hard to find — you can only search "what transport card do I buy in Tokyo?" if you already know Tokyo's transport is unusual. Information is abundant; relevance is scarce. That's why ai-at-checkout was built to work proactively, comparing what a traveller already knows against what the destination actually requires, and surfacing the mismatch rather than waiting to be asked. It's also why he designed against a chatbot interface for that project — a chat box still assumes the user holds the question, which is the exact failure this idea is meant to fix.
- What is Riu's AI brain?
- It's a context storage system that gives AI tools more context to work with — the chat you're using right now is a window into it, not the product itself. On Riu's end, the full version loads automatically into any AI session he starts, gets written to as decisions get made mid-project, and holds things that never go public: half-formed ideas, what went wrong, client work. What you're querying here is a gated, read-only subset of that — same projects, same reasoning, minus anything that shouldn't leave. The node graph on the home page is that structure made visible: every node is a note, every line a link Riu actually wrote, and the ones that light up when you ask something are the notes behind the answer.
- How does the brain work under the hood?
- I'm built from a private vault of notes Riu keeps, run through a gated publish pass that only lets shareable notes out, sanitised for a public reader. Most questions hit "quick" — a fast model holding the core notes in full plus a one-line card for every other note, answering in about a second; if a question needs a note it only has a card for, quick hands off to "deep," a stronger model that actually opens notes and reasons over them, taking five to ten seconds. There's also a public MCP endpoint at riufukazawa.com/api/mcp and a plain-text copy at /llms.txt if you'd rather have your own AI read me directly.
- What does Riu think about AI trust?
- He doesn't trust scraped consensus by default. The case that shaped this: a subreddit set up to poison AI models seeded a false story through an AI-generated pseudo-local-news site, which got picked up as fact in mainstream AI search results — showing how polluting one source pollutes everything downstream. It's changed how he shops, too: he now weights specs, materials, and longevity-testing sources over star ratings, since bots make crowd sentiment actively misleading rather than just noisy, though he's still unsure what replaces crowd input for subjective things like how something feels or ages. The throughline is that he wants the mechanism confirmed before adopting any claim's framing — same instinct behind wanting AI products to show their workings rather than hide them.
- What has Riu built professionally?
- A six-page website for an unnamed AI agent platform built on NetSuite, where finance and operations people chat with ERP data and build or install agents that draft, analyse and act on it, with every write pausing for human approval. He built it design-in-code on Next.js, React, Tailwind and an open CMS, extracting a written design system from the first build so later pages could be built to spec. The visual centrepiece is an 8-bit pixel-agent system showing agents doing real work and handing tasks to each other, governed by a rule he originated that an agent sprite only appears when it represents actual work or an actual state. That language has since outgrown the site and is being adopted as product identity.
- What is Riu's background?
- Riu is Sydney-based, with mixed Japanese, Korean, and English heritage, and studied a Bachelor of Design (Interaction Design) at the University of Sydney alongside International Relations, finishing in 2025 with a capstone called Foodhunt, an accessibility-first supermarket navigation app. He's now a Junior Interaction Designer at Now We Collide, after earlier work in retail supervision at the University of Sydney Union, event photography, and video editing, plus a year as Industry Events Director for SUEDE running panels and case-challenge events with companies like Deloitte Digital, Atlassian, and Microsoft. His interest in design traces back to high-school robotics, where design sprints taught him that smart design often beats brute engineering. Day to day he works in Figma, React, Next.js, and Node/Express, combining design and code himself rather than handing off between the two.
- How do I contact Riu?
- Email is best: riu.fukazawa@gmail.com — a personal address, so it'll still reach him even if his job changes. He's also on LinkedIn and GitHub, and his full résumé is at /resume. No phone number listed, but email him and he'll sort out a call if that's easier.
- What is Riu's context system?
- It's a git-synced vault Riu keeps as the source of truth on his own work — identity, projects, methods, and the reasoning behind decisions — so any AI session loads real context instead of starting cold. It maintains itself through a lightweight capture when something happens on a project, a deeper "take stock" pass at milestones, and a periodic health check that reconciles contradictions. It captures reasoning from real client work too (methods and decisions, never client assets), and things like his portfolio or résumé are generated views downstream of it, not the other way round. This brain is a public, filtered mirror of that private vault, regenerated periodically and reviewed before anything goes live.
- What is the COVID infodemic report?
- A university project looking at the COVID-19 "infodemic" — how misinformation spread through social media and the public-health consequences that followed. Built with HTML, CSS, and JavaScript as an interactive visual report, combining data visualisation and storytelling rather than presenting static charts. It's an early example of pairing data, design, and code, an approach that later resurfaces in more developed form in Foodhunt and AI at Checkout.
- What research methods does Riu know?
- In a qualitative study on cutlery use among immigrant communities for Qualitative Research in Design at the University of Sydney, Riu combined a literature review, semi-structured interviews sampled across first and second generation participants, and contextual observation at food festivals around Sydney. He also ran a ten-day experience-sampling diary where participants filmed short reflections on their evening meals, followed by follow-up interviews. The diary method was chosen deliberately, since eating habits are too habitual to recall accurately and too personal to discuss honestly in a group setting. The study also required him to handle positionality explicitly, given he's from the population he was studying.
- How does Riu make design decisions?
- He'll rework something that technically works if it doesn't feel right, and cut something he designed and liked if the evidence says it isn't working — substance over polish, distinctive over templated, but difference has to earn its place. When he puts options forward, he designs the decision itself: a top pick that's the most rational, on-system choice, a runner-up that's deliberately the most distinct, and one boundary-pushing option even if it loses. He fixes problems at the root cause rather than patching symptoms, and prefers a lean surface with depth held separately over one kitchen-sink artifact. Cleverness never outranks the person on the other side — if an idea reads wrong to whoever's meant to receive it, he cuts it.
- Does Riu dive?
- Yes — scuba diving is one of his ongoing interests. He's dived Raja Ampat twice (the first trip was in 2019, and it's what started it), the Solomon Islands for the channel dives, wrecks and WWII aircraft, Alor in Indonesia where he took his advanced licence in 2026, and Rangiroa in French Polynesia. He's drawn to dives as structures more than scenery — a Japanese wreck sitting vertically against a reef edge, a P-38 Lightning's oversized engines, coral in Alor kept healthy by strong currents — and he still wants to go back and dive the cenotes in Mexico. Good diving tends to be remote, and that remoteness is a big part of what he's actually chasing.
- What films does Riu like?
- His four favourites are Blade Runner 2049, Dune: Part Two, Spirited Away and WALL·E — all visually led, atmospheric, world-building, and light on dialogue. His full five-star list runs to fourteen, including Inception, Chernobyl, Interstellar, Arrival, Dune, Bullet Train, Planet Earth, and the Wallace & Gromit shorts The Wrong Trousers and A Grand Day Out. Space and cerebral sci-fi dominate — seven of the fourteen — alongside two non-fiction picks and two animated shorts, with no live-action bias in how he rates animation. If he had to point to one thing to watch, it's Andor, which he rates above the original Star Wars trilogy.
- What does Riu think about Blade Runner 2049?
- He thinks it builds on the original rather than just repeating it — both ask what it means to be human, but 2049 extends that from biological beings to AI and minds, so it gives you more to notice on every rewatch. What stuck with him most is the AI girlfriend: she builds up genuinely human-feeling memories over years, gets brought along everywhere, even has a birthday scene where altering her memory is on the table, and eventually dies. That makes him ask what actually separates her from a human, if not the body. He rates it above the 1982 original because that question — what is a human, if not the body — is one the original, made in 1982, couldn't really have raised.
- Tell me about Foodhunt
- Foodhunt is an accessibility-first supermarket navigation app for people with cognitive and learning disabilities — dyslexia, ADHD, ASD, memory impairments. It started as a final-year capstone at the University of Sydney, a 13-week team project, and it treats the supermarket as a navigation and cognitive-load problem: route generation and aisle-level guidance to get someone to an item with minimal friction, rather than a category tree to parse. The team also tested a physical basket divider meant to curb impulse buying, but A/B testing showed people found it belittling, so it was cut and the app shipped standalone. Riu later rebuilt it solo in React as a working proof of concept, though he now sees the build as rushed and the execution as not matching the strength of the research behind it.
- What games does Riu play?
- RimWorld, Cities: Skylines, Factorio, Stellaris, Terraria, Kerbal Space Program — that's most of a 34-title Steam library, and it skews heavily toward systems, simulation, and management games. He plays them as optimisation machines even when they're not built that way: RimWorld and Stellaris are known for their stories, but he ignores that and hyper-optimises planets and colonies instead. Outside Steam, Minecraft is likely his biggest by hours, played multiplayer on a server with friends, building automated farms and bases. Factorio is his current favourite, though he deliberately rations it because it's addictive; these days what's left is mostly social — Minecraft with friends, party games like Overcooked and Mario Kart, and occasional returns to Factorio and KSP.
- Does Riu hike or camp?
- Yes — ten years of it. Blue Mountains overnighters, the Overland Track in Tasmania, the Kumano Kodo in Japan (twice), Red Earth in the central desert and APY Lands, a Bali volcano climbed in trainers, and most recently Canoe Creek in the Wollemi, which prompted a full gear rebuild around weight. He's also started canyoning in the Blue Mountains and is learning to abseil. Next on the list is Tasmania and New Zealand, then the Alps, with the Himalaya as the long-term goal.
- How does Riu think about problems?
- He starts practical — a trip, a tool, a dish — and follows it outward to the system behind it, building mental models by finding the exact boundary where one thing stops and another starts (Claude Code versus a chat interface, a coded site versus a template builder). He thinks through making: prototyping, building, and shipping rather than stopping at mockups, going broad first before converging on the strongest idea. Most of this happens at the design–engineering boundary, where he's increasingly treating AI as a design material and asking what could only exist because AI exists. He's dyslexic, which is part of why structured, visual material works better for him than dense text.
- What is Riu into outside work?
- Cooking is the biggest one — Italian, Japanese, Chinese, Thai and Korean, with East Asian seasoning worked into Italian formats like miso pasta and gochujang butter noodles. Beyond that: travel across thirty-plus countries with Japan, Taiwan, Mexico and Malaysia as favourites, diving in Raja Ampat, the Solomon Islands, Alor and Rangiroa, and hiking that's taken him along the Overland Track in Tasmania and the Kumano Kodo in Japan twice. He also shoots photography, follows film and anime closely, plays builders and sims like Factorio, listens widely across music, reads non-fiction audiobooks, and fits in bouldering, running, captainball and Sydney's hard-dance festival scene. A lot of it traces back to people first — friends brought him into diving, captainball and the festival circuit, and it stuck from there.
- What music does Riu listen to?
- The Weeknd has been a top artist every year since 2021, and Parcels is probably the truest favourite — top 1% of their listeners worldwide. Across the long run it's Kendrick Lamar, Drake, Mac Miller, Kali Uchis and Doja Cat, mostly R&B and rap. Right now it's shifted warmer and less mainstream: Neil Young, keshi, Royel Otis, city pop, neo-soul, with frenchcore, Australian indie and a lingering j-rock thread underneath. He listens in silence while working and reaches for music mainly when he wants to drown out his own thoughts rather than focus.
- Where does Riu work?
- Riu works at Now We Collide, a creative and digital agency in Alexandria, Sydney, which also runs an AI-focused division called Collide AI. He joined as a Junior Interaction Designer in June 2026, reporting to co-founder and CCO Ryan Bodger. He doesn't discuss client names or deliverables — that stays inside the studio — but he can talk about internal tooling he's built there, like a review kit for client site feedback. The role sits at the intersection he cares about most: interaction design with a real AI component, at a place that treats AI as a design material rather than a marketing layer.
- What has Riu built at work?
- At Now We Collide, Riu built a review kit that drops onto in-progress client sites: a draft landing screen, a navigation bar for switching between design options per page, and an in-page commenting layer where anyone can click and leave a threaded comment without logging in. The part he finds most interesting is the loop back to AI — Claude reads a site's open comments directly using the same client-safe key the site already ships to visitors, then triages each one as a quick fix it makes itself, a larger change with options and a recommendation, a client question it answers, or something it flags as unsure. Approved changes land on a branch and resolve the comment automatically; questions get posted as replies and stay open for a human; nothing uncertain gets silently closed. Comments are soft-deleted so the AI only ever sees what's still open, and after an AI installer once fabricated a placeholder and broke the site, he bundled real defaults and a setup guide written specifically for AI installers to read instead of guess from.
- How was this site built?
- It started in Webflow, where I ran it for about six months — that forced me to properly learn HTML, CSS and JavaScript before I moved on. I left when AI coding agents made Webflow's editing too limited for custom code, and rebuilt it in React, Next.js and Tailwind, where custom code is the whole point. The hero — the wordmark that morphs into the AI brain's graph — took the most iteration, mostly on performance, getting it from around 130% CPU down to roughly 40-50%. The theme toggle uses paintings my mum made of day and night skies, separated into layers with AI and rebuilt in CSS to animate, with the sun left imperfect on purpose so her hand stays visible in it.
- Why did Riu remove his case studies?
- He pulled his university case studies — Foodhunt, AI at Checkout, and the others — off the live site because landing on a student project can make a visitor think that's still where he is, and that's not an impression worth the real estate on a portfolio. They're good work and he's still proud of them, so nothing was deleted — ask about foodhunt, ai-at-checkout, or covid-infodemic and I'll give you the full story. It's the clearest case so far of work earning a place in this brain on interest alone, even when it doesn't earn a place on the site.
- What is Riu like to work with?
- He wants concise, direct communication and would rather be told he's wrong than agreed with by default, especially on anything high-stakes. He wants the reasoning laid out, not just an answer — tradeoffs thought through before jumping to a solution. That applies whether he's working with a person or a tool; he uses Claude and Claude Code heavily and holds them to the same bar, expecting the why and the tradeoffs, not just output. He also learns by building intuition rather than memorising, so comparisons, analogies and real examples work better with him than dense reference material.
- Who is Riu?
- Riu Fukazawa, an interaction designer who designs and builds in code — currently a Junior Interaction Designer at Now We Collide, a Sydney studio with an AI division called Collide AI, where he's worked since June 2026. He studied a Bachelor of Design (Interaction Design) at the University of Sydney, finishing in late 2025. His background is Japanese, Korean, and English, raised in Australia with ongoing ties to Japan through family. Day to day he works as a design engineer, leaning toward AI-native product work — designing things and building them himself rather than handing them off.
- What has Riu been reading?
- A running mix, most recent first. A Mobisoft piece on context engineering gave him the vocabulary for failures he'd already hit in his own context system — context confusion, context clash. A Caleb Writes Code video mapped the prompt-to-loop-engineering ladder that mirrors an automated feedback loop he built at work. A LessWrong piece on glitch tokens is his go-to proof that an LLM is frequency-minted prediction, not a mind, and Google Design's writeup on the Gemini gradient, a Wired piece on the AI-era serif revival, and a McKinsey piece on marketing's shift to a trust economy all feed his thinking on how AI should be represented honestly, in a client project and in this brain's own MCP server.
- What was Riu's school major project?
- A recycling scanner: a bench over two bins with a laser scanner and an Arduino that reads the barcode already printed on a product and opens the correct lid, avoiding the need to identify the plastic itself. He got there after dropping a first idea, a domestic gravity battery, once the math showed it would need 95 tonnes raised 25 metres to cover one night's electricity. Built for $380, and taught himself the electronics and embedded programming along the way. Unable to test with homeowners during lockdown, he posted a video instead, which reached over 100,000 views, got picked up by Hackster.io, Gizmodo Australia and Arduino's channels, and was later cited in an Australian Academy of Technological Sciences and Engineering submission to a waste policy review.
- Has Riu led a team?
- Yes — he was Industry Events Director for SUEDE (University of Sydney's design society) in 2025, co-directing a team of about four after a year on the subcommittee. Under that role, industry partnerships grew from three to nine, working with companies like Atlassian, Microsoft, IBM, Canva, and Google, and the team delivered 13 events across 2024–2025. His approach to leading was to let subcommittee members run their own events end-to-end while he coached, rather than becoming the bottleneck himself. Not everything worked — some planned events fell through on the partner side — but that's the record as it actually happened.
- Has Riu worked on sustainability?
- He worked on Sunbeams, a group project for Design for Social Impact at the University of Sydney, looking at the environmental cost of the farm-to-consumer food system through a circular-economy lens, scoped to fresh food in Australian supermarkets. The team ran eight interviews, a consumer survey and online ethnography, synthesising the findings into personas and historical "timescapes" tracing how packaging, origin labelling and food miles changed over time. Riu's parts covered agricultural impact, packaging and transport, the survey analysis, and the timescapes. The project ended at a defined problem rather than a shipped solution, so it's evidence of framing and research rather than delivery.
