The Humans in the Loop is a video series exploring the people, thinking, and principles behind Tulip’s AI products — and the human-centered design philosophies shaping the future of manufacturing.
In the first episode, Tulip’s Chief Product Officer, Mason Glidden, and Product Lead, Olga Stroilova, join host Madilynn Castillo to discuss the ideas guiding Tulip’s approach to AI, and how a deep understanding of manufacturing inspired the creation of AI Composer.
The conversation doesn’t start with algorithms or interfaces. It starts with people — the engineers, operators, and designers who turn complex problems into simple, practical tools.
In an era where every company claims an AI story, Tulip’s approach begins somewhere different: on the shop floor. For Mason and Olga, the question isn’t how to add AI to manufacturing, but how to design AI that belongs there — practical, reliable, and deeply aware of context.
Mason explains that this mindset drives every decision: “If you’re not building with the people who will use the tool, you’re not really solving the problem.”
Tulip’s product philosophy — operations-first, human-centric, and open by design — means each capability is grounded in the realities of production. The team spends as much time understanding workflows, data flows, and regulatory constraints as they do on model tuning or interface design. The goal is not simply to innovate but to improve how people work and make decisions every day.
We set out to solve a real problem — that teams were spending hours re-creating information they already had.
That practical orientation shaped Tulip’s entire AI strategy. While much of the industry raced toward generative features, Tulip focused on context: how can AI make frontline operations safer, faster, and easier to evolve?
AI Composer emerged from that question — not as a grand vision, but as a response to a common frustration. In factories around the world, knowledge still lives in static documents. Operators rely on printed SOPs and binders filled with process notes. Converting them into digital systems takes weeks, sometimes months, of manual effort.
Olga describes how those conversations with customers led to the idea: “Every manufacturer we met had hundreds of PDFs they were trying to digitize. We wanted to build a way to automate that transition without sacrificing accuracy or control.”
AI Composer turns that pain point into progress. By reading structured and unstructured documents, the tool automatically generates Tulip applications that preserve the original information but make it interactive and actionable. What once lived in a binder can now trigger alerts, capture data, or connect to enterprise systems.
Still, for Tulip’s team, the focus isn’t automation for its own sake. It’s about trust. Every AI capability is designed to keep people in the loop, with clear validation steps, transparent logic, and the ability to review and modify what the model creates. As Mason puts it, “AI should make people faster and smarter, not invisible.”
Tulip’s approach to AI governance is grounded in the same principle that guides its platform: give control to the people who know the work best. Rather than relying on black-box models or rigid templates, AI Composer embeds oversight directly into the process. Users can inspect what’s generated, refine it, and apply it selectively.
That feedback loop mirrors how knowledge evolves on the shop floor. Olga compares it to an open, iterative system like Wikipedia — a living source of information shaped by its contributors. In manufacturing, she explains, that dynamic is critical: “You want AI that learns from experience, but that still has a human editor in charge.”
By balancing intelligence and intention, Tulip ensures that every AI-driven output reflects both data and expertise. The result is technology that amplifies decision-making without eroding accountability.
AI Composer also reflects a larger evolution in how manufacturers think about digitization. For years, companies sought to go paperless by recreating existing documents on screens, a common approach known as paper on glass. Tulip’s team saw an opportunity to go further, turning documentation into dynamic systems that capture data, learn, and improve.
When AI Composer transforms a PDF into a Tulip app, it doesn’t just replicate the instructions — it connects them to real-time context. Operators can log performance data, supervisors can trace quality metrics, and engineers can update processes instantly across sites. What was once a static artifact becomes a living part of continuous improvement.
This isn’t just technical progress; it’s cultural. The tool changes how teams think about knowledge itself — from something stored and retrieved to something shared and evolved.
AI in manufacturing has to be composable, governed, and human-in-the-loop. People need to see it working before they trust it.
Both Mason and Olga return to the same idea throughout the episode: Tulip’s advantage isn’t its algorithms; it’s its empathy. Every feature is shaped through dialogue with real users — operations leaders, line engineers, and digital transformation teams who understand the complexity of their environments better than anyone else.
That understanding shapes every design choice, from interface simplicity to validation frameworks. AI Composer isn’t just built for manufacturers; it’s built with them, through shared testing, feedback sessions, and live pilots that turn customer insights into design patterns.
The result is a product that feels familiar to the people who use it — because they helped define it.
As the conversation turns to what’s next, Mason and Olga are optimistic but measured. The future, they agree, won’t be defined by a single breakthrough model but by the ability to combine AI with human creativity in safe, scalable ways.
Multimodal inputs, agentic workflows, and generative authoring will all have their place, but the principle remains the same: keep people in the loop.
Mason sums it up simply: “It’s really about how you give that human superpowers — help them do the thing they’re uniquely good at, and give them leverage on everything else.”
That balance between human judgment and intelligent tools defines Tulip’s vision for the future of manufacturing. In an industry built on precision, safety, and skill, the most powerful transformation is still the most human one: technology that helps people think, act, and build better every day.
🎥 Watch the full series of The Humans in the Loop on YouTube.
