Dragoon

i organized everything i've written on crypto design into one page. brand strategy, case studies, wallets, onboarding, defi, frameworks, mental models. not a blog archive. a curriculum ↘︎ designprotocol.substack.com/p/the-library currently free access for everyone, paid plan soon.

dropping thursday ↘︎ phantom won by making wallets pretty. @Rabby_io @DeBankDeFi won by making wallets honest. different thesis. both work. breaking down how rabby thinks about design for power users. explore more in the meantime ↘︎ designprotocol.substack.com

Agents, meet the Figma canvas

#58 built forgeui as an experiment.

I rather take 1000 shots where I try something new and fall on my face in front of everyone vs be the guy that has 1000 ideas and never goes after any of them.

#57 built glow for a friend who runs a small salon. she was managing bookings across texts and calls, walk-ins on memory, and inventory by guessing when things ran out. watched her work for a day. clients waiting without knowing how long. staff double-booked because the schedule lived in three places. repeat clients slipping through because nobody tracked when they were due back. built a system that handles all of it. calendar with drag-and-drop scheduling, walk-in queue with auto-assignment that actually explains why it picked that slot, inventory that deducts automatically when services complete, revisit reminders based on service type. added ai chat so she can ask things like "who's overdue for a visit" without digging through screens. still adding things based on what she actually needs running the shop.

Designers are not cooked. We’re still the chefs and they can’t stand it.

#56 built onesheet because i got tired of sending decks that were already out of date by the time anyone opened them. decks are static. you make them, send them, then spend the next call explaining what changed since you exported the pdf. metrics are stale. roadmaps shifted. the deck becomes a lie the moment you close the editor. built a single-screen canvas instead. company name, what you do, live metrics with definitions, roadmap, explicit risks, and a proof stream of actual shipped artifacts. everything structured. everything inspectable. one url that stays current because it's the source of truth, not a snapshot of one. the core idea is working: a company should be legible in one screen, and that screen should be honest.

#55 built mindtrace out of a question: what if agents could understand how you decide, not just what you produced? every ai tool gets your outputs. documents, messages, code. but the reasoning that led there - the options you rejected, where you hesitated, what constraints shaped your thinking - that context disappears. you have to re-explain yourself every time. built a decision capture system that runs entirely inside the app. compare options, track dwell time, log choices and rejections, rate outcomes. turns all of that into a live graph, inferred preferences, and agent-ready packets you can paste anywhere. no external tracking. no background monitoring. just structured decision moments you create yourself. local-first, shareable via url. still fixing render bugs and refining the feed system. but the core idea is working: capture enough decisions and the tool starts showing you patterns you didn't notice yourself. experimenting with whether portable context changes how agents respond to you.

launch announcement ↘︎ introducing "the protocol" → design patterns and strategic insights for building better crypto products and brands. the only publication breaking down design, brand strategy, and creative direction for crypto as business infrastructure, not decoration. until now, this was invite-only. shared privately with my circle. today, it's open for everyone. If you found it useful, share it with your friends or rt. help me get this in front of the right people. here's the full story ↘︎ i've been working and obsessing over crypto brand, product, design and creative strategy for years. somewhere between the curiosity and the work, i started curating, organizing and writing everything down. breakdowns. frameworks. patterns. mistakes. things i wished someone told me earlier. never meant to build a publication. just kept curating and writing. put it on substack, ran it private, shared it with a few founders i trusted. then a creative block hit. i backed the whole thing up and deleted it. classic. came back eventually. reorganized everything. and realized - nobody else is writing about design, creative direction and brand strategy for crypto as business infrastructure, not decoration.` so here we are. same work, proper home. if you're a founder, vc, or designer in web3 - this is for you. welcome to the protocol:) ↓ read it here designprotocol.substack.com

if you can imagine it, you can build it

#54 built archive out of curiosity. wanted to understand what happens between saving inspiration and actually using it. every tool in this space is a filing cabinet. save images, organize into folders, then figure out the direction yourself. the gap between collecting and executing is still manual. built a moodboard-to-direction engine. save 5+ references, hit extract direction, get a color palette with assigned roles, tag clusters showing the visual dna, a written brief summarizing what the collection is saying, and exportable css/json tokens ready to paste into a project. still experimenting with the extraction logic.

#53 building something called shipmuse: an ai app builder that takes prompts or screenshots and outputs working code with live preview. the prompt-to-app space is crowded now. lovable, bolt, v0, replit. but i wanted to understand the pipeline myself. how do you turn vague english into a structured spec? how do you make codegen deterministic? how do you handle the model returning garbage json? built an appspec system that converts prompts into a validated schema first, then generates a file plan, then writes code in chunks with validation between each step. screenshot-to-app uses vision to extract layout, colors, and components before generating. local-first with no signup. bring your own api key. still fixing things. template gallery outputs were collapsing to the same result. file operations weren't applying correctly. the kind of bugs you only find when you actually use the thing. not sure if this becomes a product or just teaches me how these tools work under the hood. building it either way.

imp announcement ↘︎ gm gm folks, surprise, this was @0xDragoon the whole time. crypto x ai native designer, 2x founder, worked w/10+ crypto teams, helped founders raise millions, led + scaled startups, ai x vc x tech. early @0xPolygon follow me → @0xDragoon, that's where i drop work. explorations. ideas. builds. and more. if anything i've posted ever helped you in any way, rt this. it costs you nothing and it means everything to someone who's building. that's it. back to posting. thanks:)

Play is work's only moat

#52 built a tool... Patternflow: an svg pattern designer for creating repeating tiles, dot grids, and exportable pattern systems. built tool while working on a client brand project.still adding things. still experimenting. but the core loop already works better than what i was doing before.

the most underrated hire right now is a great product person. when i say product person i'm def not talking about a product manager. perhaps i think there has to be somewhat of a new role. i don't have a good name for it yet but maybe something like "product thinker".. someone with an intuitive grasp of the product as it exists, where it's soft, where it sings, & how to iterate it toward something even sharper. in some sense, this person has to cohesively hold in their head where this product should be 2 years from now & work backwards from that. i say this cuz when building was hard, engineering was the bottleneck & the status hierarchy often reflected that. building is no longer hard. which means the variance in outcomes has shifted almost entirely to judgment on what to build, how to sequence it, & how to talk about it. & the story matters as much as the thing. internally, it organizes the team around a shared model of why. externally, it shapes the interpretive frame users bring to their first experience. you can't retrofit narrative onto a product & expect it to land, it has to be load bearing from the start. the rarest version of this person sits at the intersection of culture & deep technology. someone genuinely bilingual. they know what's technically possible & they know which cultural currents are real vs. ephemeral. that combo is what separates products that feel inevitable from products that feel assembled. before ppl clap back with this person has always been valuable, i know.. i am just saying now they might be the most *important* person in the room. their value compounds like never before.

#51 built an app... ClinicCue: a lightweight front desk system for small clinics that just works. i built cliniccue for my friend. one screen that handles the entire front desk: live token queue, appointments, billing with inventory tracking, patient notes with medical templates, and follow-up reminders with whatsapp shortcuts. she opened it once and never went back to the notebook.