How Four Startups Harness OpenAI to Change Everyday Work

The pace of change in AI never fails to surprise me, even when, as someone fascinated by automation and the nitty-gritty of digital workflows, I think I’ve seen it all. Recently, OpenAI DevDay 2025 brought together a band of daring startups who don’t just talk about AI—they’ve woven it tightly into the fabric of their daily work. During a real-time broadcast of the OpenAI Podcast, founders from Cursor, SchoolAI, Abridge, and Jam.dev sat down to share how OpenAI tools are helping them reshape their respective industries.

AI used to be something of a “club”—for coders, data scientists, and the odd tech writer with a penchant for late nights and strong coffee. Nowadays, I see its fingerprints everywhere: in classrooms, hospitals, codebases, and in teams squabbling (politely, of course) over bug reports. With these four companies as my guides, I had a chance to peek behind the curtain and discover how OpenAI technology has gone from theoretical to, well, indispensable.

A Glimpse into the Everyday Magic of AI

For this article, let me walk you through the insights shared by these founders during the podcast. Their experiences are a masterclass in bringing AI “down to earth”—turning it from a high-minded ideal into a genuine ally at school, work, and even the doctor’s office.

• SchoolAI – Revolutionising classroom experiences for teachers and students alike.
• Jam.dev – Lightening the load for software teams by automating bug reporting and QA.
• Abridge – Translating the complex world of medical documentation into plain language.
• Cursor – Bringing AI to coding, from suggestion to automation.

Let me introduce each and show how they’re not just riding the AI wave—they’re teaching it new tricks.

SchoolAI: A Personal Touch for Every Lesson

The sight of students hunched over exercises or teachers juggling marking with lesson planning is familiar to almost anyone. When I heard Caleb Hicks from SchoolAI describe their solution, I couldn’t help but think how much easier my own tutoring days might have been with such a tool at my disposal.

Personalised Content at Scale

Imagine a platform that lets teachers whip up interactive lessons on the fly, automatically generating quizzes, exercises, and even feedback—all tailored to each student’s needs. SchoolAI, with the help of OpenAI’s models, builds content that actually responds to learners’ strengths and weaknesses.

This isn’t just automatic marking (though I’d have killed to have that myself, back when I was buried under stacks of essays). What makes SchoolAI stand out is its knack for picking up on subtle cues—adapting to the style of a teacher or the learning journey of a student with what I can only call digital intuition.

• Automated feedback and individualised assignments
• Resources that adapt over time, based on both teacher and student input
• Real-time analysis to help teachers spot students at risk of falling behind

From what Caleb shared, it’s clear that SchoolAI isn’t “replacing” educators—it’s bolstering their superpowers. In a world where so many kids risk falling through the cracks, that’s no small feat.

Balancing the Human and the Digital

Admittedly, technology in the classroom can be a double-edged sword. I’ve seen plenty of so-called “smart” platforms that feel coldly generic, more focused on metrics than on learning. SchoolAI seems to get that teaching is as much about relationships as results. By making lesson creation less of a slog, it gives educators back their most precious resource: time. I’d venture a guess many teachers are secretly toasting to that.

Jam.dev: From Chasing Bugs to Meaningful Work

If you’ve never seen a software team argue over a bug report, count yourself blessed. It’s rarely glamorous—endless screenshots, cryptic logs, and a fair bit of wishing you were anywhere else. I nearly laughed when Dani Grant from Jam.dev described how their AI assistant swoops in to take on the grunt work, leaving developers free to tackle problems that actually require creativity.

No More “Needle in a Haystack”

Jam.dev takes a refreshing approach to software testing and quality assurance. Here’s what caught my attention:

• Automated log analysis to pinpoint root causes of errors
• Generation of concise, actionable bug reports, complete with context and reproduction steps
• Continuous monitoring to spot patterns and “silent failures” before they snowball

The result? Less time lost in the weeds, and more spent creating features users actually want. When bug-hunting shifts from a manual slog to an automated flow, team morale tends to soar—trust me, I’ve been there.

Empowering the Creative Side of Development

Dani stressed how freeing up engineers from repetitive QA means each moment spent working is, well, actually productive. A bit of time for blue-sky thinking, a few fewer headaches during code reviews—a quiet revolution in how teams collaborate.

Jam.dev’s approach makes it easier to iron out snags early, keeping projects “on the rails” rather than descending into last-minute panic. I can only imagine how many cups of emergency coffee have been saved.

Abridge: Cutting Through Medical Jargon

Here’s a confession: I still sometimes leave the doctor’s office with a swirl of half-remembered instructions and a small mountain of paperwork. Listening to Zach Lipton from Abridge, I found myself nodding in agreement—how handy it would be if someone could decode all that medicalese for me.

Understanding Healthcare, One Note at a Time

Abridge leans on OpenAI technology to:

• Extract key insights from doctor-patient conversations
• Summarise notes and highlight actionable items for both patients and clinicians
• Translate jargon into plain English, bridging the gap between complex terminology and everyday understanding

This helps both sides of the healthcare table. Clinicians save precious minutes (and possibly their sanity), while patients gain clarity—a rare commodity in a system known for its opacity.

From Drowning in Documents to Concise Communication

Zach painted a picture of a world where critical follow-ups don’t fall through the cracks, simply because patients or doctors missed a detail. I’ve experienced the anxiety that comes from not quite understanding what’s next—so a tool that lifts the fog is, in my book, a quiet champion.

Abridge serves as something of a personal health translator, levelling the playing field and giving both patients and providers a fighting chance at clear communication.

Cursor: The Future of Coding, Now More Accessible

There’s a certain magic to telling a computer what you want, and watching it figure out how to get there. When Lee Robinson shared Cursor’s approach, I realised just how much coding has changed since the days I was banging out HTML by hand.

Enter Coding Agents

Cursor deploys what Lee calls “coding agents”—AI-driven assistants that can:

• Generate, refactor, and test code snippets with minimal input
• Automate repetitive programming tasks
• Work alongside humans, offering suggestions or even tackling tickets independently

This isn’t just about boosting productivity (though, let’s be honest, who doesn’t want that?). Cursor uses tools like AgentKit and Apps SDK from OpenAI, making it possible for both seasoned coders and novices to build complex automations—even if they’ve never heard the term “dependency injection.”

A New Philosophy of Software Creation

Lee described a workflow where developers simply describe what they want to accomplish—and the agent figures out the rest. It’s as if the IDE has learned to read minds and roll up its metaphorical sleeves.

Having spent my fair share of time debugging mystery errors, I can appreciate just how liberating that could be. It’s a shift from rote typing to problem solving, from command lines to conversations. We’re no longer so much sharing desks with computers, as instructing (and correcting!) a particularly industrious junior colleague.

Highlights from OpenAI DevDay 2025

Tech events can sometimes be all smoke and mirrors, but DevDay 2025 struck me as refreshingly hands-on. OpenAI introduced several tools that grabbed my attention—and looked well-suited to the needs of both startups and large teams alike.

• AgentKit – A visual agent builder, tools for monitoring, and methods to evaluate AI’s decisions in real time.
• Apps SDK – Rapid building and launching of apps inside ChatGPT, blurring the line between chatbot and true assistant.
• New models such as GPT-5 Pro and gpt-realtime-mini – faster, more affordable, and ready for integration into everyday platforms.
• Plug-ins and integrations with popular services like Booking.com, Canva, Figma, Expedia, or Spotify—bringing AI into apps people actually use every day.

For me, AI becomes truly valuable when it weaves itself invisibly into familiar workflows. This goes well beyond tech demos and enters the realm of practical, everyday value.

Beyond Hype: Concrete Impact in the Real World

One thing stood out to me as I listened to these startup founders: We’ve moved past the stage of AI “experiments.” The technology has found its way into the very core of how work is done—quietly, gradually, but with unmistakable impact.

• Teachers find themselves with more time and resources to connect with students.
• Software teams swap tedium for creative problem-solving.
• Clinicians and patients navigate the maze of medical information with far less confusion.
• Developers spend more time designing solutions than wrangling code by hand.

In each case, the theme is clear: AI isn’t about shiny gadgets or sci-fi dreams. It’s about quietly making life and work a bit less tangled.

What Sets These Startups Apart?

If you’ve slogged through “innovation” pitches in the past, you’ll know it’s easy for a startup founder to get lost in their own buzzwords. Not here. These four companies share a few habits worth highlighting:

• Practicality First: Each tool addresses stubborn, everyday pain points.
• Empathy: Whether it’s a harried teacher or a puzzled patient, the user’s real needs always come first.
• Adaptation, not just automation: The AI adapts to people—not the other way around.
• Collaboration: Humans and machines work together, not in opposition.

During the podcast, the founders spoke less about replacing jobs and more about making expertise accessible. I caught a strong sense that the best AI doesn’t push people aside; it lends them a (virtual) hand, letting them spend time on the parts of the job that truly matter.

The Road Ahead: Everyday AI, Not Just for the Experts

I grew up in the era when computers meant arcane commands and frequent headaches. Watching AI become as commonplace as a coffee break is, in its own quiet way, a bit miraculous. These startups aren’t just building flashy apps; they’re quietly weaving smart tools into the warp and weft of daily life. As OpenAI’s new models and dev tools reach more hands, I expect the trend to accelerate.

For teams like those behind Cursor, SchoolAI, Abridge, and Jam.dev, the real measure of success isn’t how clever the AI is, but how fluidly it bends to human needs. You feel the difference the first time a machine saves you from mindless paperwork or helps you untangle a stubborn bug. The delight isn’t in the technology alone—it’s in the unexpected minute gained, the frustration avoided, the student who finally “gets it.”

Why This Matters—For Startups and the Rest of Us

When I look ahead, I can’t help but think the winning formula will involve a blend of accessibility, adaptability, and practical utility. Tools like those discussed at DevDay 2025 are dragging AI out of the domain of experts, into every workplace, classroom, and surgery.

A word of caution, though: seamless as it might appear, this wave of AI does ask something from each of us. It calls for new habits, openness to change, and a bit of patience when the bots (inevitably) get a detail wrong. But as the saying goes, “no rose without thorns”—the adjustments are a small price to pay for the freedom these tools can provide.

Getting Started: Bringing OpenAI Into Your Own Workflow

If you’re reading this and wondering how you might start benefiting from OpenAI-powered tools, the good news is you don’t have to be a seasoned developer or a tech wizard. The new resources announced at DevDay—AgentKit, Apps SDK, expanded plugins—are lowering barriers across the board.

• For educators: Start by experimenting with SchoolAI’s interactive content generators or automated feedback. Customise to your class, tweak as you go—see how the platform learns with you.
• For developers: Give Cursor a spin. Try letting an agent refactor a stubborn chunk of code, or integrate an automation for testing that’s been eating your evenings.
• For healthcare professionals: Check out how Abridge could save time on notes and help bridge gaps in understanding for your patients.
• For anyone juggling bug reports: Jam.dev might just give your team room to think, by handling repetitive QA and root cause analysis automatically.

In my experience, the first task you hand off to an AI assistant can feel a little like magic. Over time, though, it just becomes another reliable part of the day’s routine—like your favourite mug or a well-worn shortcut key.

Final Thoughts – Practical AI, Real-World Results

As I step back from the wealth of stories and observations shared during the OpenAI DevDay Podcast, one thing comes into sharp focus: innovation is at its best when it fades into the background, quietly improving everyday work. The startups presented here serve as a proof point for those still on the fence about AI in their own professional world.

It’s never just about the latest model or the cleverest prompt, but about genuine problems solved and minutes reclaimed. These tools don’t just save time—they open up space for the very human magic of learning, invention, care, and creative thinking.

If you’re ready to make your own leap, don’t wait for AI to become perfect. Test something new, welcome the odd hiccup, and celebrate every small win as you go. These four startups offer a living blueprint for what’s possible—and, sometimes, that’s all the encouragement you need to get started.

Fancy diving into the full conversation? The OpenAI Podcast episode from DevDay 2025 is available for listening, alongside other thought-provoking content. With a bit of luck, after tuning in, you’ll spot new ways to make your own daily routine a little smarter, too.