Why I Gave a Voice to AI at the Anthony Nolan Buildathon

I’m Dmytro Zuiev, a full-stack developer passionate about using technology where it matters most - not just for scale, speed, or sales, but for impact.

In November 2024, I joined the AI for All Buildathon at 42 London. It was a two-day sprint filled with energy, vision, and urgency. My team worked with Anthony Nolan, a UK charity that saves lives by matching stem cell donors with people battling blood cancer. Our mission: build an AI-powered chatbot that could transform how they support donors - and free up more time for life-saving work.

Anthony Nolan AI Chatbot interface

Why AI, Why Now?

Every week, Anthony Nolan receives 150+ emails and calls from potential donors asking questions like:

“Am I eligible to donate?” “What happens after I register?” “How do I update my details?”

Most of these answers already exist — buried in web pages or PDFs. But when someone is taking a step toward saving a life, the last thing they should feel is confused or alone.

So we built Felix, an AI assistant that doesn’t just answer questions - it reassures, guides, and frees up the charity’s time to do the work that matters most.

Why I Joined This Project

As someone who’s changed careers, traveled the world, and supported people from all walks of life, I know how hard it can be to navigate uncertainty. That’s why I’m drawn to projects where technology becomes invisible, and the person in need comes first.

Anthony Nolan is not just another charity. They believe in compassion, clarity, and courage. And those beliefs shaped how we built the bot — not just as a tool, but as an extension of the care their team provides every day.

What We Built in 48 Hours

It wasn’t just a chatbot - it was a fully integrated AI-powered assistant, designed for real users with real questions. Here's what we pulled off in just two days:

  • Next.js 15 frontend with a modular, responsive chat UI
  • Anthropic & Langtail AI models for context-aware, safe answers
  • Dynamic message handling with live streaming using Vercel AI SDK
  • FAQ integration via custom Langtail variables
  • Bing-powered semantic search tool for real-time site queries
  • Feedback mechanism to ask users if their questions were answered
  • Auto-resizing, embeddable widget with postMessage communication to allow seamless integration into the charity’s site
const result = await streamText({
  model: langtail('chatbot', {
    environment: 'production',
    variables: { faqTemplates: JSON.stringify(faqTemplates) },
  }),
  messages,
  maxTokens: 1024,
  temperature: 0.7,
  maxSteps: 2,
  tools: {
    bing_search: tool({
      description: 'Search for Anthony Nolan related information',
      parameters: z.object({ query: z.string() }),
      execute: async ({ query }) => {
        const searchResults = await formattedBingSearchResults(query)
        return searchResults
      },
    }),
  },
})

The assistant pulls content directly from Anthony Nolan’s website, adapting its tone and response to the user’s intent - and making sure people get the right help, fast.

AI That Frees Up Human Time

The real win? With Felix answering common questions, the Anthony Nolan team can focus on things that no AI can do: providing emotional support, solving complex cases, and recruiting more stem cell donors - potentially saving more lives.

We didn’t just write code. We created a second voice for a life-saving organization, one that works 24/7, never gets tired, and always points people in the right direction.

What This Taught Me

  • That AI has soul - when built with empathy and purpose
  • That a chatbot can be more than helpful - it can be life-changing
  • That the best software doesn’t replace people - it amplifies their impact

For me, this project was a reminder of why I chose this path. Not just to build cool things, but to build useful, human things.

If you’re working on a mission that matters - in health, social impact, or beyond - and need a developer who thinks about both the code and the people it serves, I’d love to collaborate.

Because when we get the tech right, we give people back something even more valuable:

Time. Confidence. Hope.