Today's Workflow Tools Are For Programmers

ORCFLO began as a passion project in 2025, consuming our nights and weekends...becoming a real company in 2026. All this happened in an unplanned 'stealth mode'. This series explains the back-story of our journey.

By this point in our journey, we had a clear understanding of what was wrong with chat interfaces and what we wanted instead. We wanted agentic workflows - structured sequences of AI tasks that we could design, execute, monitor, and improve over time. We wanted to orchestrate agents together in workflows...the name ORCFLO emerges!

So we went looking for tools that would let us build these workflows. The AI market is hot, everyone is building everything, and everyone talks about 'Agentic' all the time - how hard could this be?

We scanned the market carefully. We tried platforms, watched demos, read documentation. And we kept running into the same problem.

Everything was built for developers.

The interfaces assumed familiarity with code. The setup processes required technical configuration. Even the "no-code" and "low-code" options, when we dug into them, turned out to require a fair bit of heavy-lifting - technical configuration, managing API keys, etc. We're tech guys and CAN do all of this, but we just need a simple solution that doesn't demand coding for every new situation. We didn't want to wire together a radio, we just want to play some music.

This was frustrating, but it wasn't surprising. The AI space has been driven by technologists, and they've naturally built tools for themselves. That makes sense. But it also means that a huge population of potential users are locked out. And by 'users', we mean people like us - those who understand business problems in-depth and have complex problems that could be solved using agentic workflows.

What we wanted was simple: a way to build agentic workflows without programming. Something visual and intuitive. Something that let us focus on what we're good at - defining the work to be done - rather than struggling with how to make the technology cooperate.

We couldn't find such a solution in the market.

So the question became: could we build it ourselves?

NEXT UP:
Step 6: From Tinkering to Launch