# Supabase Projects at AI Tinkerers

> Canonical HTML: https://aitinkerers.org/technologies/supabase
> Markdown URL: https://aitinkerers.org/technologies/supabase.md
> Technology record last updated: 2026-09-18T15:13:58Z
> Generated: 2026-09-21T19:48:13Z

The open-source PostgreSQL development platform: a Firebase alternative for rapid backend deployment.

Supabase is the Postgres development platform, providing a complete, open-source backend-as-a-service solution. It packages enterprise-grade tools like a full PostgreSQL database, a RESTful API (via PostgREST), and a GraphQL API (via pg_graphql) that are auto-generated from your schema. The platform includes a comprehensive suite of services: Auth (for user sign-ups and SSO), Storage (for files with S3 integration), Realtime (for database change subscriptions), and Edge Functions (for serverless logic). The core value proposition is clear: build fast with a powerful, scalable SQL database that offers 100% portability and robust security features like Row Level Security (RLS).

- Official technology site: https://supabase.com
- Public AI Tinkerers demos and talks: 90
- Result page: 1 of 4

## Recent Public Talks and Demos

### [AI for Feels, Not Just Tasks](https://la.aitinkerers.org/talks/rsvp_d3x3AU8Ehzk)

Glad to present how I'm going about building this emotional gps for AI - basically a tool that helps you get emotional clarity in 5 minutes. It's not therapy - it's in between group chat and therapy. Users type in what's hard to say. The app using AI shows images for visual projection to surface emotional patterns, and then it reflects back what could really be going on, what you actually want and what you're prepared to do about it. Happy to present a demo of the working product, then go into the file architecture, my workflow and unit economics for API calls.

- Event context: Aug 16 - AI Tinkerers LA: Beauty, Bytes, and Venice Beach Vibes 🌊✨🤖 — Los Angeles
- Public talk page: https://la.aitinkerers.org/talks/rsvp_d3x3AU8Ehzk

### [Eliluce : Du dessin à la créature : donner vie à l’imagination des enfants avec l’IA](https://paris.aitinkerers.org/talks/rsvp_jueOAUu0c6w)

J’ai créé Elliluce, une application créative qui permet aux enfants de dessiner des créatures imaginaires, puis de les transformer en personnages illustrés grâce à l’IA. Pendant la démo, je dessinerai une créature dans l’application, je lancerai sa transformation et je dévoilerai le résultat à côté du dessin original. Je montrerai également comment le dessin est transmis au modèle pour générer l’illustration.

- Event context: AI Engineer Kids Day — Paris (Ages 9–15) — 2026-09-20 — Paris
- Public talk page: https://paris.aitinkerers.org/talks/rsvp_jueOAUu0c6w

### [Selling AI Agents With AI Agents](https://boston.aitinkerers.org/talks/rsvp_CoVeIXBfmhM)

I'm a solo founder building around a day job, so the constraint is hours, not ideas. I built an outbound GTM engine that runs on a Mac Mini at my desk and runs the pipeline while I'm at work: weekly lead sourcing and qualification against a narrow ICP, personalized first-touch outreach, revival sequences for dead leads, cadenced follow-up for warm leads, and a Slack command center that flags replies, requests approvals, and tees up the sales calls worth taking. It also learns from its own results: every reply, every silence, and every booked call feeds back into which signals get sourced next and how the next round of outreach is written, one variable at a time. The engine doesn't care what it's selling. Swap the ICP and the signals and the rest holds. My first test case was an AI quoting agent for screen printing shops, where the ICP was owner-operated shops with an automatic press and an existing quote form. The same design rule runs through all of it: an LLM reads the messy inputs, deterministic rules make every decision that has to be right, and a human takes the calls that matter.

- Event context: Back from Summer: AI GTM Builders — 2026-09-03 — Boston
- Public talk page: https://boston.aitinkerers.org/talks/rsvp_CoVeIXBfmhM

### [From Meow to Mastery: Building MathCat’s Adaptive Learning Platform](https://hong-kong.aitinkerers.org/talks/rsvp_nk7cD0uzBpI)

MathCat.fun is an adaptive, bilingual math and science practice platform for Hong Kong primary school children, aligned to the EDB curriculum and playable as a Progressive Web App. For this demo, I’ll walk through the full stack, not just the game UI. Live, I’ll briefly show the working app (different play modes, leaderboard, rewards shop, etc); but will focus more on the backend: (1) the question-generation and validation workflow – the balance between deterministic and probabilistic models to achieve the best outcome and my learnings; (2) talk about the AI critic pipeline (Gemini + GPT-5) that scores each question template on Cantonese naturalness, HK school terminology, hint quality, and age fit (3) the adaptive progression architecture—mastery gates, recent-accuracy windows, skill unlocks, etc; (4) the overall architecture.

- Event context: AI Tinkerers Hong Kong: August Meetup with OAX Foundation and GMAsia — 2026-08-31 — Hong Kong
- Public talk page: https://hong-kong.aitinkerers.org/talks/rsvp_nk7cD0uzBpI

### [Measuring velocity direction in AI Era](https://islamabad-rawalpindi.aitinkerers.org/talks/rsvp_a5xgW0Yy5VY)

Momentum; A velocity measuring tool that helps in driving teams in right direction with the right velocity in AI Era. ## THE DEMO A real internal system that we're using and still building to measure velocity, direction and overall organization's momentum. We'll talk about how each matric is being captured, how multiple matrices are complimenting each other and how various team roles i.e. admin, team lead or individual member can read them.

- Event context: AI Tinkerers Islamabad - August 15, 2026 — 2026-08-15 — Islamabad Rawalpindi
- Public talk page: https://islamabad-rawalpindi.aitinkerers.org/talks/rsvp_a5xgW0Yy5VY

### [\[Case Study\] 90,000 rejections, 100 meetings: the AI outbound engine I optimized to not send](https://valencia.aitinkerers.org/talks/rsvp_9Hq5aZKj4fk)

An AI-run cold outbound engine for a link-building agency — a CRM I own end to end. It sources prospects, enriches them, finds their contact data automatically, scores them 0–5 against our ICP, then writes personalized multichannel sequences (email + LinkedIn) built from each prospect's website, LinkedIn profile, and recent activity — and hands them to a sender, while every reply flows back into one channel-agnostic thread model I can query in plain language. To date: ~257k prospects processed, 175k messages generated, 1,000+ replies, 100+ sales meetings, and six figures of revenue directly attributable. The counterintuitive part — and what I'll show live — is that the machine is optimized to reject, not to convert. Every send spends finite domain and sender reputation, so between each stage sits a gate whose job is to throw work away: the score gate permanently drops ~90k non-fit prospects before writing a word, and a send-time re-check has blocked 1,591 already-approved messages. Live I'll walk the pipeline dashboard and per-job run history, the score distribution and its cutoff, real generated messages (a good one and one that was embarrassingly wrong), and the per-prospect cost ledger — 1.64M rows — that prices every decision.

- Event context: GTM Teardown August — AI Tinkerers Valencia — 2026-08-11 — Valencia
- Public talk page: https://valencia.aitinkerers.org/talks/rsvp_9Hq5aZKj4fk

### [Protecting Secrets from Long-running Agents](https://boston.aitinkerers.org/talks/rsvp_qu4vU02kztE)

I built Sesame which is an auth broker proxy for AI agents that protects your secrets from never getting exposed to the agents. Sesame brokers the request and protects you from any unintended actions. Sesame offers full monitoring and alerting mechanisms so human operators can make safe agentic mutation especially in production environments.

- Event context: AI Tinkerers Boston: GTM Agentic AI Launch — 2026-06-29 — Boston
- Public talk page: https://boston.aitinkerers.org/talks/rsvp_qu4vU02kztE

### [Three Models, One Ledger: A Live Multi-Model Witness Protocol over MCP](https://austin.aitinkerers.org/talks/rsvp_NEzGU02MWI8)

A three-model witness protocol: to verify a piece of AI-assisted work, the same proof is sent to Claude, GPT, and Gemini independently over MCP, and it only seals into an append-only ledger when they agree. Live, I'll fire a witness request from my phone, show all three models attest in real time, and watch the record seal into the vault — the payloads, the disagreement handling, and the actual database row. No slides.

- Event context: AI Tinkerers Austin x PostHog Demo Night — 2026-06-18 — Austin
- Public talk page: https://austin.aitinkerers.org/talks/rsvp_NEzGU02MWI8

### [Orquestração multi-agente em n8n: sub-workflows virando tools dinâmicas](https://curitiba.aitinkerers.org/talks/rsvp_F61Msr-AenQ)

Construí um sistema multi-agente que opera o assistente de WhatsApp de uma marca brasileira de suplementos "clean label" com catálogo de 20+ SKUs em 8 categorias (sono, estética, emagrecimento, imunidade, saúde feminina 40+, entre outros) e dois canais ativos de WhatsApp para quiosque físico e delivery. O problema real: volume de perguntas simultâneas sobre produtos específicos (composição, indicação, contraindicação, disponibilidade), mix de intenções radicalmente diferentes: consumidora final comprando pela primeira vez, cliente em recompra, nutricionista prescrevendo e a necessidade de escalar atendimento sem perder o tom consultivo que a marca construiu. Em vez de LangGraph ou CrewAI, usei n8n como orquestrador e Claude como motor de decisão. Mensagens chegam via Evolution API num webhook do n8n; o Claude classifica a intenção (tool use) e roteia para a tool certa; mas cada "tool" é um sub-workflow n8n separado, chamado via Execute Workflow. Cada sub-workflow tem seu próprio prompt, seu próprio acesso ao Supabase (RAG sobre o catálogo completo com pgvector, histórico de pedidos, perfil do cliente) e seu próprio guard-rail. Quando a confiança do Claude cai especialmente em perguntas clínicas sobre contraindicações, o agente passa para humano via Chatwoot. Tudo self-hosted na Hetzner (Docker Swarm + Traefik). Na demo ao vivo mostro: uma mensagem real chegando no WhatsApp, o painel do n8n executando nó a nó, o sub-workflow correto ativando como tool, e os logs no Supabase. Compartilho o padrão "sub-workflow-as-tool" e o JSON exportado.

- Event context: AI Tinkerers Curitiba: Evento Inaugural — 2026-06-10 — Curitiba
- Public talk page: https://curitiba.aitinkerers.org/talks/rsvp_F61Msr-AenQ

### [Website RAG-based chatbot that doesn't suck (much)](https://seattle.aitinkerers.org/talks/rsvp_3o5HjPc5kJo)

"What if visitors to our website could chat with our years of free content without making sh*t up?" Will show the bot itself live, walk through how it works, and talk about the future integration.

- Event context: GTM / Growth Track - Seattle — 2026-05-27 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_3o5HjPc5kJo

### [Revolutionalizing audiobooks with AI](https://dubai.aitinkerers.org/talks/rsvp_WNyvbI0-h8Q)

75000+ Audiobooks narrated to you through the power of AI. Will start with completely explaining the product along with a working demo of the app. A user chosen book (from the 75000+) book catalogue being narrated live with AI.

- Event context: AI Tinkerers Dubai - May Demo Day — 2026-05-23 — Dubai
- Public talk page: https://dubai.aitinkerers.org/talks/rsvp_WNyvbI0-h8Q

### [GTM in the Terminal — Agent-driven personalized outbound](https://sf.aitinkerers.org/talks/rsvp_O2H7CQUJ6IU)

Cyrus is the cofounder of The Synthesis Company (YC S24). Previously, he was a core contributor to Stanford NLP's DSPy framework and cofounder of WOMBO.ai (100M+ downloads). He'll be demoing GTM-OS, a CLI and TUI he used to programmatically drive hyper-personalized LinkedIn outbound. For AI tinkerers he built usesocial.dev, a CLI to connect your agents to LinkedIn / X and build your own workflows and systems.

- Event context: AI Tinkerers San Francisco: GTM Engineering Track — 2026-05-19 — San Francisco
- Public talk page: https://sf.aitinkerers.org/talks/rsvp_O2H7CQUJ6IU

### [ICM - Interpreted Context Methodology](https://columbus.aitinkerers.org/talks/rsvp__HgYuH0-pAY)

I've built upon Jake Van Clief's Interpretable Context Methodology (ICM) - to write my multi-step agent workflows as folders with markdown inside of them. Skill files to get started are linked.

- Event context: AI Tinkerers - Columbus May Meetup — 2026-05-04 — Columbus
- Public talk page: https://columbus.aitinkerers.org/talks/rsvp__HgYuH0-pAY

### [Merging Multi Agent Council Delibrations with Dynamic Analytics to Redesign Transit Infrastructure](https://toronto.aitinkerers.org/talks/rsvp_zWQT1Z51PV8)

A sandbox planning tool where you draw new Toronto transit lines and a 6-agent AI council critiques it; with the ability to see real-time changes and analytics for routes and ridership as the line is developed.

- Event context: AI Tinkerers Toronto - April 2026 - hosted by Shopify — 2026-04-29 — Toronto
- Public talk page: https://toronto.aitinkerers.org/talks/rsvp_zWQT1Z51PV8

### [From Messy Inputs to Reliable Community Data: Human-in-the-Loop AI](https://denver-boulder.aitinkerers.org/talks/rsvp_EuM3A-aBwJ4)

I built Colorado Songwriters Collective, a human-in-the-loop AI workflow that converts messy event/member inputs (DMs, forms, and web pages) into structured, publish-ready community data. In the demo, I’ll show the ingestion pipeline, review checkpoints, and how we keep updates accurate before they go live.

- Event context: AI Tinkerers Denver - April Meeting — 2026-04-27 — Denver
- Public talk page: https://denver-boulder.aitinkerers.org/talks/rsvp_EuM3A-aBwJ4

### [Your GTM Copy Only Reaches 32% of Buyers — Here's How to Measure It](https://seattle.aitinkerers.org/talks/rsvp_UuJIGqJNdf0)

COS is a content measurement engine that scores how well B2B sales and marketing copy resonates across all five buyer personality types (Big Five/OCEAN). It orchestrates 7 parallel analysis frameworks through a 24-module knowledge base (~38K lines of psychology research) to surface which buyer segments you're reaching and which you're missing. Live at cos.semalytics.com, free guest mode, no signup.

- Event context: AI Tinkerers Seattle: GTM Engineering — April Meetup — 2026-04-23 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_UuJIGqJNdf0

### [Supercharging Development with Claude Cowork](https://montreal.aitinkerers.org/talks/rsvp_73dcPJFKniM)

Coluur is a watercolor paint matcher for iOS and web — upload a photo and it returns the exact Sennelier, Daniel Smith, Winsor & Newton, or Holbein tubes (with pigment codes, lightfastness, and transparency ratings) that reproduce every colour in it, plus two-paint mix recipes when no single tube is close enough. Demo plan: I'll upload a reference photo live and walk through the extraction and match results (~30s), then hand the screen to Cowork — Anthropic's desktop agent — to have it pick up a real Notion ticket from the Coluur sprint board and ship a visible change to the running app in under a minute.

- Event context: AI Tinkerers Montreal - April Demo Night — 2026-04-22 — Montreal
- Public talk page: https://montreal.aitinkerers.org/talks/rsvp_73dcPJFKniM

### [Postal - AI Chief of Staff that keeps me organized and updates itself](https://austin.aitinkerers.org/talks/rsvp_lQCh4jkzo7o)

Postal is the OpenClaw for people that don't want to buy a Macbook. I love the power of AI for coding and information discovery, and I wanted the same for my day-to-day tasks, like writing emails, updating calendar, keep track of notes, and memories about people and places. Postal offers all of that in one, convenient interface, available on my desktop and my phone. Web site is Supabase realtime, desktop Tauri runs completely offline with its own filesystem. Local models can be plugged in to stay off grid. The sync between environments is still janky, but I am working on it. The agents do research, reminders, rewrite my own notes, triage emails... the board is slowly eating its own dogfood and improving itself. Currently the app managing 900 different "memories" for me.

- Event context: AI Tinkerers Austin: April 2026 Demo Night — 2026-04-16 — Austin
- Public talk page: https://austin.aitinkerers.org/talks/rsvp_lQCh4jkzo7o

### [Claude Code for Patent Prosecution](https://cologne.aitinkerers.org/talks/rsvp_0141nYwU_RQ)

"Claude Code for Patent Prosecution," features an agent designed to support and automate the prosecution of patents at the European Patent Office (EPO). The demo showcases the end-to-end pipeline: from a highly reliable, exact ingestion of dense technical PDF documents, through an AI review process constrained by strict EPO guidelines, all the way to generating native .docx files with "track changes" enabled for attorney review.

- Event context: AI Tinkerers Cologne 4: Live Technical Demos — 2026-04-16 — Cologne
- Public talk page: https://cologne.aitinkerers.org/talks/rsvp_0141nYwU_RQ

### [Connecting Chicago engineers with the right companies](https://chicago.aitinkerers.org/talks/rsvp_TXG7B89_Hys)

A hyperlocal talent marketplace for engineers

- Event context: AI Tinkerers Chicago: April Meetup ft OneTwoLoop — 2026-04-14 — Chicago
- Public talk page: https://chicago.aitinkerers.org/talks/rsvp_TXG7B89_Hys

### [What Happens When You Put an AI Dev Team on a Mac Mini?](https://chicago.aitinkerers.org/talks/rsvp_lY6IQju70NA)

I built a local AI agent development team on a Mac Mini to support software development for HausHavn, a real estate workflow platform. Instead of using AI for one-off prompts, I set up separate agents with clear roles across architecture, coding, QA, product analysis, documentation, and project management. The demo shows how I orchestrate those agents through real development work, including task handoffs, code review, GitHub issues, Notion sync, release QA, and the messy lessons of making agents useful as process infrastructure, not magic.

- Event context: AI Tinkerers Chicago: April Meetup ft OneTwoLoop — 2026-04-14 — Chicago
- Public talk page: https://chicago.aitinkerers.org/talks/rsvp_lY6IQju70NA

### [TriCoach](https://boston.aitinkerers.org/talks/rsvp_HM-11Iww8A0)

I built an app to help me train for my triathlon. It provides a daily dashboard with everything I need to do, along with an AI coach that adapts to my progress, needs, and effort levels.

- Event context: AI Tinkerers Boston: March 2026 Meetup — 2026-03-30 — Boston
- Public talk page: https://boston.aitinkerers.org/talks/rsvp_HM-11Iww8A0

### [Solving Cold Start GTM with Claude Code (and Zero Headcount)](https://seattle.aitinkerers.org/talks/rsvp_saEaMuqlges)

Loupe is a luxury watch market intelligence and tool platform designed for collectors and dealers. With a one-person team and Claude Code, we ship SEO landing pages, Instagram content (carousels, reels, stories), editorial guides, across 22+ watch brands, all from CC. In this demo, I'll walk through the actual workflows: how a single CLI session can research a topic, query Supabase for watch data, generate templated visuals with Puppeteer, compose ffmpeg reels with Ken Burns and transition effects, and push a finished landing page and associated Instagram reel to production. Leveraging Claude Code with custom slash commands, custom MCP integrations, and a growing skills library that turns repetitive GTM work into repeatable pipelines. I'll show the real session transcripts, the messy iterations, and the parts that broke. Leveraging this method, we've grown from 0 to 100 followers in a week.

- Event context: AI Tinkerers Seattle: GTM Track — March — 2026-03-26 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_saEaMuqlges

### [Ai in Human Resource](https://bremen.aitinkerers.org/talks/rsvp_M5plVNmMjOw)

An HRMS software where I am using an LLM for context-based answers.

- Event context: AI Tinkerers Bremen — 2026-03-25 — Bremen
- Public talk page: https://bremen.aitinkerers.org/talks/rsvp_M5plVNmMjOw

## Related Technologies

- [Next](https://aitinkerers.org/technologies/next) ([Markdown](https://aitinkerers.org/technologies/next.md)) — 186 public demos
- [React](https://aitinkerers.org/technologies/react) ([Markdown](https://aitinkerers.org/technologies/react.md)) — 220 public demos
- [TypeScript](https://aitinkerers.org/technologies/typescript) ([Markdown](https://aitinkerers.org/technologies/typescript.md)) — 205 public demos
- [Claude Code](https://aitinkerers.org/technologies/claude-code) ([Markdown](https://aitinkerers.org/technologies/claude-code.md)) — 215 public demos
- [Vercel](https://aitinkerers.org/technologies/vercel) ([Markdown](https://aitinkerers.org/technologies/vercel.md)) — 48 public demos
- [Claude](https://aitinkerers.org/technologies/claude) ([Markdown](https://aitinkerers.org/technologies/claude.md)) — 174 public demos
- [Gemini](https://aitinkerers.org/technologies/gemini) ([Markdown](https://aitinkerers.org/technologies/gemini.md)) — 188 public demos
- [OpenAI API](https://aitinkerers.org/technologies/openai-api) ([Markdown](https://aitinkerers.org/technologies/openai-api.md)) — 520 public demos
- [FastAPI](https://aitinkerers.org/technologies/fastapi) ([Markdown](https://aitinkerers.org/technologies/fastapi.md)) — 181 public demos
- [n8n](https://aitinkerers.org/technologies/n8n) ([Markdown](https://aitinkerers.org/technologies/n8n.md)) — 46 public demos
- [Python](https://aitinkerers.org/technologies/python) ([Markdown](https://aitinkerers.org/technologies/python.md)) — 662 public demos
- [ElevenLabs](https://aitinkerers.org/technologies/elevenlabs) ([Markdown](https://aitinkerers.org/technologies/elevenlabs.md)) — 42 public demos
- [Anthropic API](https://aitinkerers.org/technologies/anthropic-api) ([Markdown](https://aitinkerers.org/technologies/anthropic-api.md)) — 66 public demos
- [Cursor](https://aitinkerers.org/technologies/cursor) ([Markdown](https://aitinkerers.org/technologies/cursor.md)) — 65 public demos
- [Node](https://aitinkerers.org/technologies/node) ([Markdown](https://aitinkerers.org/technologies/node.md)) — 99 public demos
- [Codex](https://aitinkerers.org/technologies/codex) ([Markdown](https://aitinkerers.org/technologies/codex.md)) — 44 public demos
- [Docker](https://aitinkerers.org/technologies/docker) ([Markdown](https://aitinkerers.org/technologies/docker.md)) — 147 public demos
- [GPT-5](https://aitinkerers.org/technologies/gpt-5) ([Markdown](https://aitinkerers.org/technologies/gpt-5.md)) — 30 public demos

## More Results

- Next: https://aitinkerers.org/technologies/supabase.md?page=2