Crypto startup
DamFi
Built and sold a crypto product company. Owned product direction, roadmap, operations, marketing, team coordination, investor conversations, analytics, and execution.
Build notesAvailable for startup roles
I build products.I bend AI into tools.I keep shipping.
Founder engineer, AI native product builder, and startup operator in Los Angeles. I built and sold a crypto startup. I built LineCite, a legal AI product that failed commercially but made me much sharper. Now I’m building TextTea, SetClipper, and Layance while doing standup comedy.
I’m useful when the job is vague, the product is half real, the deadline is rude, and someone needs to turn the fog into software.
Built and sold a startupShips consumer AI appsMobile plus backend plus AIFounder level ownershipProduct judgment under chaos
Built for weird startup problems, not committee theater.
Selected Work
Selected work I can defend in public.
Built products, sold a startup, shipped apps, failed loudly enough to learn, and kept going. No fake founder mythology. Just receipts and scar tissue.
Consumer AI
TextTea
AI app for decoding text conversations, tone, subtext, mixed signals, and emotional chaos. Built, shipped, launched, and iterated from blank repo to App Store.
Agentic spend
Layance
Agentic spending governance for AI agents and modern business spend. Controls what agents can spend, who they can pay, how much, under what approval rules, and with what audit trail.
Creator tools
SetClipper
AI workflow for comedians and podcasters. Turns raw recordings into searchable bits, laugh moments, clip candidates, and performance history.
Legal AI
LineCite
Citation backed legal AI product for medical chronology and litigation prep. Commercially failed. Useful scars in trust, evaluation, provenance, and product rigor.
Build Notes
Build notes from shipped and failed products.
A few products I built, launched, sold, or learned from the hard way. No mythology. Just what happened, what I made, and what it shows.
Crypto startup
DamFi
Summary
Crypto product company built and sold.
Role
Founder and CEO.
What I built
I built and operated a crypto product company from early concept through sale. I owned product direction, roadmap, execution, operations, marketing, team coordination, investor conversations, and analytics.
Why it mattered
DamFi shows I can do more than build prototypes. I can operate inside ambiguity, push a product forward, coordinate people, make tradeoffs, and drive toward a business outcome.
What this shows
I can think like a founder, execute across functions, and stay accountable for outcomes beyond the codebase.
Stack and workflow: Crypto product operations, product strategy, Python data workflows, analytics, marketing systems, team coordination, investor materials, roadmap planning.
Live consumer AI app
TextTea
Summary
Consumer AI app for decoding text conversations, tone, subtext, mixed signals, and emotional chaos.
Role
Founder, product builder, mobile engineer, release owner.
What I built
I took TextTea from product idea to live iOS app. I owned the React Native and Expo app, Firebase backend, subscriptions, App Store release flow, product copy, onboarding, screenshots, QA, and launch iteration.
Why it mattered
TextTea shows I can ship a real consumer product, not just prototype something in a sandbox. It required product taste, mobile execution, App Store work, payment flow, attribution, and fast iteration.
What this shows
I can own a vague product from blank repo to shipped app and handle the ugly middle where product, engineering, copy, QA, and launch all overlap.
Stack and workflow: React Native, Expo, Firebase, Firestore, EAS, Adapty, AppsFlyer, App Store Connect, Codex, Claude Code, Gemini.
In progress
Layance
Summary
Agentic spending governance for AI agents and modern business spend.
Role
Founder, product strategist, systems designer, fintech builder.
What I built
I am designing the control layer for agent driven spending: what agents can spend, who they can pay, how much they can spend, which approvals are required, and what audit trail exists after the fact.
Why it mattered
As AI agents become more capable, businesses need controls before they let software take financial actions. Layance is about permissions, policies, approvals, vendor controls, auditability, and trust.
What this shows
I can think through emerging infrastructure problems where product, trust, risk, and engineering all collide.
Stack and workflow: Product architecture, fintech workflow design, policy systems, approval flows, audit trails, spending rules, agent tool permissions, early stage technical planning.
In progress
SetClipper
Summary
AI workflow for comedians and podcasters that turns raw recordings into searchable bits, laugh moments, clip candidates, and performance history.
Role
Founder, product designer, AI workflow architect, full stack builder.
What I built
I designed the core product objects: sets, bits, moments, clips, transcripts, recordings, and performance history. The system is built around extracting useful creative structure from messy recordings.
Why it mattered
SetClipper shows I can design product systems for an emotionally specific user, not just generic dashboards. It combines video understanding, creator tooling, search, organization, and workflow design.
What this shows
I can turn an ambiguous creative workflow into a structured product with clear objects, actions, and user value.
Stack and workflow: React, backend workflows, transcript processing, Gemini video analysis, model evaluation, structured extraction, Codex and Claude Code engineering loops.
Built and sunset
LineCite
Summary
Citation backed legal AI product for medical chronology and litigation prep.
Role
Founder, product engineer, AI systems architect.
What I built
I built a legal AI product focused on deterministic evidence extraction, citation backed claims, medical chronologies, provenance, evaluation gates, and attorney ready outputs.
Why it mattered
LineCite failed commercially, but technically it made me much sharper. It forced me to care about truth, traceability, single source of truth, evaluation, regression tests, and not letting AI invent facts.
What this shows
I know how to build AI systems where correctness matters. I do not treat AI output as magic. I build gates, tests, provenance, and review loops.
Stack and workflow: Python, OCR pipelines, evidence graphs, extraction pipelines, evaluation harnesses, regression gates, canonical event modeling, citation verification, deterministic rendering.
Why people hire me
Product instinct.
Systems brain.
Shipping reflex.
I’m not the cleanest box on an org chart. That is the point. I’m best where product, engineering, copy, tooling, release quality, and founder judgment all collide.
0 to 1 without adult supervision
I can take a vague product idea and turn it into flows, code, copy, tests, releases, and a working thing people can actually judge.
AI native by default
Codex, Claude Code, Gemini, Cursor, OpenRouter, and Ollama are part of my daily build loop. I use them to prototype, test, debug, compare models, and move faster without turning the codebase into soup.
Founder range
Product judgment, engineering execution, launch, analytics, app store work, subscriptions, attribution, customer research, positioning, and the little haunted details that make products real.
Systems brain
I like messy domains where truth matters: evidence graphs, extraction pipelines, evaluation harnesses, spending controls, search, ingestion, and workflows with consequences.
Background
One startup sold.
One startup failed.
Both made me better.
I’ve spent my career building systems across defense, crypto, legal AI, and consumer apps. I’m strongest in early stage environments where the product is still becoming itself and the work does not fit neatly into one job title.
I can think like a founder, ship like an engineer, write like a product person, and use AI tools like extra limbs.
2018
DIA
Data systems and intelligence workflows
2020
Metronome
Software for Navy data integration
2023
DamFi
Crypto startup built and sold
2025
LineCite
Legal AI built and sunset
2026
TextTea
Consumer AI app shipped
Now
Layance
Agentic spending governance
Let’s talk.
If you need someone who can turn ambiguity into shipped product, I’m interested.
Especially if the role involves AI, fintech, consumer apps, developer tools, founder chaos, product design, or building the first version of something that does not fully exist yet.
paddytobagen@gmail.com
Los Angeles, CA
linecite.com
Available for product, engineering, AI prototyping, startup operator, and founding engineer roles.
1000x engineer cyborg. Mostly housebroken.