Saint Marlo Labs turns raw ideas into polished web and mobile apps — from Lovable prototypes to GitHub-backed codebases, live domains, and App Store launches.
A growing portfolio of products across travel, events, health, touring intelligence, podcasting, productivity, and AI workflows.
The group chat for coordinated travel.
An AI-powered group travel and event coordination app that brings chats, plans, places, polls, payments, calendars, media, and responsibilities into one shared space.
Touring intelligence from ticket links, screenshots, and seat maps.
A touring intelligence product that monitors ticketing pages, seat maps, pricing, on-sale performance, market conflicts, and promoter-ready reporting.
ChravelApp is the flagship build: an AI-powered group travel and event coordination app designed to replace chaotic group chats with one shared operating system for plans, places, people, payments, and decisions.
Every Saint Marlo build runs through custom agent harnesses, per-project prompt libraries, and a rotating fleet of frontier models — Codex, Google Jules, Claude Code, Lovable, Replit, Cursor, and GitHub — chosen per task to play to each model's strengths.
We study the top coding signal daily — Twitter, Reddit, Google Developer accounts, Anthropic and OpenAI release notes — and fine-tune every prompt, agent, and harness against the latest model behavior. It's a craft, not a shortcut.
Daily review of frontier-model updates, prompt research, and tooling drops from Twitter, Reddit, Google Developer accounts, Anthropic, OpenAI, and the Cursor / Codex / Jules / Replit changelogs. Anything worth keeping is folded into the harness inside a week.
Per-project agent rigs — planner, coder, reviewer, QA — versioned and iterated on continuously. Not one-shot prompts, not vibes. Real scaffolding tuned per codebase.
Project-specific prompt packs maintained per app and re-tuned weekly against new GPT, Codex, Claude, and Gemini behavior. When a model shifts, the prompts shift with it.
Codex, Google Jules, Claude Code, Lovable, Replit, Cursor, and GitHub all in rotation. Each task is routed to the model that wins on it — Claude for refactors, Codex for long-horizon agents, Lovable for UI velocity, Cursor for surgical edits, Jules for async PR work.
GitHub-backed source of truth, Supabase / Lovable Cloud, auth, RLS, edge cases, perf budgets (INP < 200ms, LCP < 2.5s), and regression-free diffs. No demos shipped as products.
Domain, PWA, TestFlight, App Store, monitoring, and ongoing operation — not just launch day. Every product stays in the harness after release.
Build receipts, not buzzwords. Every Saint Marlo Labs project is backed by real repos, real commits, and real iteration loops — but the product story stays curated.
The group chat for coordinated travel.
Real-time discovery for what's happening around you.
Modern clinic and wellness stack management.
Repo link coming soon
AI strategy and intelligence for podcasts.
Each project moves through the same shipping ladder — from live web app to PWA, then TestFlight, submission, and public launch.
Saint Marlo Labs is building a portfolio of AI-native apps across consumer, professional, creator, health, travel, and touring workflows.