Standardize processes, ensure food safety, and manage all your locations from a single platform. Checklists, audits, tasks, campaigns, and much more.
Side projects, all of them early. Softerize is still the main work — these are smaller bets, each with its stage marked on the card. Further down, the smaller tools I have published as open source.
Agentic chat for your whole team, in a Slack-style app. Connect any LLM with an API key or through the CLI you already use (Claude Code, Codex), give each agent a role and tools, approve every write before it happens, and share the agents with your team on a hub you host.
Why three projects
Your project context — endpoints, env vars, commands and encrypted secrets — in typed #packs, one command away from your shell with the tsk CLI. Plus a keyboard-first todo list.
Why three projects
An indie studio making cozy, goofy games for the whole family. The first ones are in the oven.
Why three projectsTools I wrote (or an AI wrote) to solve a problem of my own and published because it is probably not only mine. Open source, free, and with the issues open.
A two-pane editor that converts markdown ⇄ HTML through the clipboard. Paste from Word, Google Docs or an AI chat and get clean markdown — and go the other way to paste already formatted into an e-mail.
Read the articleI started three projects in 2026: Comitiva, ShellTasks and Abyssurd. What each one is, what stage it is at, and why none of them changes where Softerize stands.
Almost every markdown⇄HTML converter swaps text for tags. Markdown Bridge renders the other side, and that turns Ctrl+C/Ctrl+V into the whole solution — from Word, from ChatGPT, or from any page.
The real cost of a decision is not making it, but fixing it. When being wrong is expensive, people stop deciding and the whole company slows down. On flexible software, a culture of continuous deployment, and why AI exposes who has already solved this.
An analysis of the 2026 Restaurant Technology Outlook report. How operational efficiency, data and AI will protect margins and the P&L in foodservice.
We live in an age of urgency. Why email is still the king of productivity, and how the 'WhatsAppification' of work is draining your energy and fragmenting communication.
In the fast-moving startup ecosystem, the choice of technology stack can be the difference between a successful launch and an endless analysis loop. Building our own SaaS, we embraced the Monolith First philosophy — in a contemporary form that avoids the headaches of legacy monoliths.
The restaurant sector is in a curious moment: economic uncertainty on one side, aggressive expansion on the other. An analysis of Crunchtime's 2025 Restaurant Growth Insights Report and its lessons for the Brazilian market.
Is it worth predicting the future? In software, yes. But be careful not to lock up the innocent. In this article I write about the 'Minority Report Solution' in systems design, and how to prevent the fatal error without getting in the user's way.
How do you guarantee that the taste, the plating and the food safety at your restaurant in São Paulo are exactly the same as at a location in Manaus? For foodservice chains, this is the daily challenge.
For almost a decade, we at Softerize worked with a total focus on engineering. Plenty of development, plenty of features and... very little marketing. Now, in 2026, we are starting a new stage.
Master of Information Technology Management (MITM)
The University of Sydney
Australia
February, 2011 - November, 2011
MBA in Project Management
Fundação Getúlio Vargas (FGV)
Brazil
March, 2006 - September, 2007
Computer Science
Universidade Federal do Rio Grande do Sul (UFRGS)
Brazil
March, 2000 - January, 2005
Softerize
CEO & Founder
Brazil
September, 2009 - present
Ziller
Web Programmer
Australia
May, 2011 - January, 2012
SAP Brazil
Solution Developer
Brazil
August, 2006 - August, 2009
eFact Software
Systems Analyst
Brazil
March, 2006 - July, 2009
ADP Brazil
Developer
Brazil
April, 2005 - February, 2006
Produttare Consultores Associados
Intern
Brazil
September, 2003 - January, 2005
