Introduction
Healthcare professionals often have brilliant ideas for improving patient outcomes but lack the coding knowledge to bring those ideas to life. Enter Arun, a London-based NHS pharmacist, Clinical Entrepreneur, and self-described technophile who found himself repeatedly stymied at hackathons—always needing a developer’s help to turn clinical insights into tangible solutions. Everything changed when he discovered Lovable (formerly GPT Engineer), an AI-driven, no-code platform that unlocked the potential to build full-featured apps through simple text prompts. Arun has since created over 20 applications, organized Clinical AI Hackathons across the UK, and is even venturing into quantum computing.
TL;DR
- Arun’s Background: A London pharmacist who joined hackathons but felt held back by coding limitations.
- Discovering Lovable: Found GPT Engineer (now Lovable) and began building full-featured healthcare apps through simple prompts.
- Major Projects: Created the QOF Progress Tracker, Smart Blood Test Assistant, and a Patient Triage Navigator—no coding experience required.
- Hackathon Wave: Arun organized Clinical AI Hackathons, drawing 300+ participants and spanning multiple UK locations.
- Quantum Twist: Incorporated quantum computing seminars into hackathons, collaborating with companies like Quantinuum.
- Mistral’s Le Chat Copy: Used a Lovable starter template to swap the OpenAI integration with Mistral’s API in just a few prompts.
Early Days & Discovering Lovable
S: Hi Arun, can you tell us a bit about your background and how you first stumbled upon Lovable?
A: Sure! I’m Arun, a pharmacist based in London and I starting working in 2012. I’ve always been fascinated by technology, so I’d join these hackathons where developers and clinicians would team up to brainstorm healthcare solutions. But the biggest hurdle for me was coding—I had none of those skills. Fast-forward to September of last year: I was part of a decentralized science (DeSci) fellowship in London called Muse Matrix. During some online research and discussions with fellow clinicians, I discovered a tool called Bolt that helps you build apps. But it only let me do three prompts a day.
Then someone said, “Hey, try Lovable!” Back then, it was known as GPT Engineer. I hopped onto Lovable, and suddenly I got 100 prompts free. That was huge—Bolt had only given me three. Lovable let me build entire websites and apps by just chatting with the AI. I didn’t need to rely on a developer anymore, and that changed everything.
From Pharmacist to App Builder
S: You mentioned you’ve built more than 20 apps—could you share a few examples that stand out?
A: Absolutely. One of them is the QOF Progress Tracker, which helps GP practices monitor Quality and Outcomes Framework (QOF) performance. Another is the Smart Blood Test Assistant, which suggests appropriate blood tests based on patient conditions. We also have a Patient Triage Navigator, an intelligent system that directs patients to the right care pathway quickly. In all these projects, Lovable’s AI engine did most of the heavy lifting.
Interestingly, I don’t dive into the code. I prompt Lovable with instructions like “Hey, connect to this external API,” or “Add a card to display patient information,” and it just does it. Over time, I started to pick up on certain coding terms and best practices, but I never had to become a full-fledged programmer.
Hackathon Movement
S: So you’re not only building apps, you’re organizing hackathons. How did that come about?
A: It all started when I kept seeing fellow NHS clinicians—doctors, nurses, pharmacists, even social prescribers—who had brilliant ideas but lacked coding skills. After my early successes with Lovable, I thought, “Why not organize a hackathon specifically for frontline healthcare workers?” I mentioned the idea in my NHS Clinical Entrepreneur group and in my British Tamil Entrepreneurs circles as I am French Tamil.
To be honest, I expected maybe 10 people to sign up. But after we posted a quick registration page (built with Lovable, by the way) and linked it to Resend’s API, we got 80 sign-ups in six hours, and over 100 by the end of the day. I had to pay for Resend’s membership just to handle all the email captures! Interest exploded from there. We now have over 300 registrants and seven different locations for the hackathon across the UK.
Merging AI & Quantum: A Bold Experiment
S: You mentioned quantum computing—how does that fit into a healthcare hackathon?
A: Funny story. Right after announcing our “Clinical AI Hackathon,” I discovered an IBM Quantum event happening in London. I was curious, so I went, and the big takeaway was that quantum computing and healthcare might be a match made in heaven—things like optimization problems, drug discovery, or advanced analytics can potentially benefit from quantum solutions down the line.
I had the idea to rebrand to “Clinical AI & Quantum Hackathon.” I got some pushback—people said clinicians wouldn’t understand quantum. Even I didn’t know much about it! But I connected with a company called “Quantinuum,” which agreed to join and give a basic presentation on quantum, especially for healthcare applications. It ended up being such a hit that we’re hosting our grand finale at Quantinuum’s headquarters in a couple of months.
I’ve also enrolled in quantum courses through the National Quantum Computing Centre (NQCC) in the UK who have a partnership with Q-CTRL which is an Australian company. I pulled some late nights, studying so I could guide hackathon teams better. It’s intense, but I love seeing these out-of-the-box solutions clinicians come up with when they’re not bound by traditional software or hardware constraints.
Copying Mistral’s “Le Chat”
S: You also mentioned you cloned Mistral’s “Le Chat” interface by repurposing a Lovable starter template. How did that work?
A: It was surprisingly simple. Lovable has a ready-to-go template for a chat app that usually uses OpenAI’s API in the backend. I said, “Change the integration to Mistral’s API,” and I pasted in my key. With a prompt or two, the chat was running on Mistral instead of OpenAI. I swapped out logos, tweaked the design, and—boom—had a working “Le Chat” clone. I think, in total, I spent maybe 2–3 prompts on that particular build. It took me a few more to fine-tune the layout, but it was shockingly quick.
I was more intrigued by the speed at which Lovable let me replicate functionality from one API to another. It’s a testament to how flexible the platform is. Now I do it for my friends and hackathon participants if they want to switch out one large language model for another.
Diving Into the Technical Side
S: You’ve clearly gone deep into Lovable’s features. Which integrations or APIs do you find most helpful for healthcare?
A: We often use:
- Resend for sign-up forms and emailing participants.
- Supabase for quick database solutions.
- Stripe if we ever need payment functionality, though we haven’t used it much for the hackathons.
- Stability AI for image-related tasks like generating medical diagrams.
- ElevenLabs for AI-powered voice generation—some people love to do voice-based interfaces.
- Mapbox for location-based features in apps, like finding nearby clinics or tracking medical deliveries.
- Replicate and Runware for more advanced AI or generative features.
Building an Ecosystem of Clinician-Coders
S: You seem to have become a mentor figure—coaching other clinicians to build their first apps. How do you train them on Lovable?
A: I usually have them start with a simple Product Requirements Document (PRD). They write out exactly what they want the app to do. Then we break it down into prompts. For example:
- “Create a basic webpage with sign-up functionality.”
- “Connect to Supabase for data storage.”
- “Add a chat feature using Mistral’s API.”
- “Design a user-friendly layout with a custom logo.”
Within two or three prompts, Lovable typically gets them a functional MVP. After that, they can refine design elements, add advanced features, or tweak the user interface. We often spend a couple of hours together, especially if they need multiple integrations. But honestly, it’s a short learning curve because Lovable does the coding.
If you’re stuck, you can always open another AI chat (like Claude) for debugging tips. But usually, Lovable has you covered. The main advice is: Don’t try to do everything in one monster prompt. Break it down step by step, and keep building from there.
The Future of Healthcare Tech
S: You’ve seen frontline challenges firsthand. Where do you think healthcare technology—especially no-code and AI—will lead us in the next few years?
A: We’re in the middle of a real transformation. I see AI-driven tools, from triage to risk assessment, becoming standard in all settings—hospitals, GP practices, community pharmacies, even dentists. One big trend is that clinicians themselves will build these solutions instead of waiting for large healthtech companies. That’s the essence of democratizing innovation.
I’m also fascinated by quantum computing’s potential for drug discovery and complex optimization tasks. Looking further, there’s VR/AR for surgical planning, drone delivery for high-cost or emergency medications, and 3D printing personalized pills. The NHS is historically cautious about adopting new tech, but the success of these hackathons shows there’s massive enthusiasm on the ground level. Once we break down coding barriers, the sky’s the limit.
Conclusion
Arun’s story is a powerful testament: with the right AI-driven platform, even those without traditional programming skills can create sophisticated tools that address frontline healthcare needs. From his first small experiments with Lovable to orchestrating large-scale Clinical AI Hackathons, Arun proves that passion, creativity, and an open mind can catalyze real change.
He doesn’t see himself as just a pharmacist anymore—he’s part developer, part innovator, and part educator for his colleagues. In a healthcare landscape that can feel slow to adapt, Arun’s approach shows a path forward: empower the people closest to the problems to build the solutions.
Ready to explore how Lovable can turn your ideas into functional apps—no code required? Jump in and start creating. You might just spark a revolution in your own industry.
