I’m excited to share some news that means a lot to me: Llamatik has been selected as one of the five winners of the 2026 Kotlin Foundation Grants Program, receiving a $6,000 grant to support its continued development.
For an independent open-source project that started about a year and a half ago, this is an incredible milestone. More importantly, it’s an opportunity to take Llamatik further and continue exploring what AI can become when it runs directly on our devices.
What is Llamatik?
For those discovering the project for the first time, Llamatik is an open-source Kotlin Multiplatform ecosystem designed to make running large language models (LLMs) directly on devices easier for developers.
Built around Kotlin Multiplatform and native inference technologies such as llama.cpp, Llamatik provides tools and SDKs for integrating AI capabilities into applications across Android, iOS, Desktop, and Web.
The idea behind Llamatik is simple: AI shouldn’t always require an internet connection, cloud infrastructure, or expensive API subscriptions.
Running AI locally offers several advantages:
- Privacy: Data can remain on the user’s device instead of being sent to external servers.
- Offline availability: AI-powered features can work without a continuous internet connection.
- Accessibility: Developers can experiment with AI without depending entirely on commercial cloud APIs.
- Sustainability: Making better use of existing hardware and reducing unnecessary cloud dependency can contribute to more efficient AI applications.
Llamatik aims to make these capabilities accessible to Kotlin developers through a multiplatform, developer-friendly ecosystem. If you want to try it right away, my guide on running LLMs offline on Android with Kotlin is a good starting point.
Building Open Source Between Work and Family
Llamatik started as a personal experiment driven by my interest in Kotlin Multiplatform, native development, and the possibilities of local AI.
Over time, it gradually evolved into something much more ambitious.
But building and maintaining an open-source project alongside a full-time job and family life comes with its challenges.
With a wife and two kids, the time available for Llamatik is limited. Most development happens during those few hours left at the end of the day, whenever I can find a moment between professional and family responsibilities.
There are days when I make significant progress, others when I spend hours debugging a single issue, and plenty when I simply don’t have time to work on the project.
Progress isn’t always as fast as I’d like, especially when there are so many ideas I’d love to explore.
Nevertheless, the opportunity to learn, experiment, and create something useful for other developers makes the effort worthwhile.
That’s one of the reasons this recognition from the Kotlin Foundation means so much to me.
It’s not just financial support. It’s a validation of the project’s potential and an encouragement to keep building.
What the Kotlin Foundation Grant Means for Llamatik
The $6,000 grant will help support the next stages of Llamatik’s development, particularly in areas that can make the ecosystem more useful and accessible to developers.
Some of the main priorities include:
Improving the SDK and Developer Experience
One of my main goals is to make integrating local AI into Kotlin Multiplatform applications as straightforward as possible.
This means improving the SDK architecture, simplifying APIs, expanding documentation, and providing practical examples that developers can use in their applications.
Expanding Multiplatform Capabilities
Llamatik aims to provide a consistent developer experience across different platforms while taking advantage of their native capabilities.
Improving compatibility, performance, and platform integration will remain an important part of its evolution.
Building Local AI Agents
Running language models locally is only the beginning.
I’m particularly interested in building AI agents capable of performing tasks, interacting with applications, and executing workflows directly on users’ devices.
The goal is to give developers the foundations for creating intelligent applications that can operate independently of cloud-based AI services whenever possible.
Looking Beyond a Single Device
My longer-term vision for Llamatik goes beyond running an individual language model on a phone or computer.
Imagine an ecosystem where multiple devices can discover each other, share their available computing resources, and collaborate to perform AI-powered tasks.
An old laptop, a smartphone, or a desktop computer could potentially become part of a distributed network of local AI capabilities.
There are still significant technical challenges to overcome, and this remains an exploratory direction rather than a finished feature.
But I believe there’s an interesting future in combining on-device inference, intelligent agents, and distributed computing.
Llamatik is my way of exploring that future while keeping the technology open and accessible.
A Thank You to the Kotlin Community
I want to sincerely thank the Kotlin Foundation for selecting Llamatik and supporting independent open-source development.
Recognition like this makes a real difference, particularly for projects maintained by individual developers in their spare time.
I’m also grateful to everyone who has explored Llamatik, contributed to the project, reported issues, or simply shown interest in its development.
And congratulations to the other recipients of the 2026 Kotlin Foundation Grants Program!
What’s Next?
There’s still a lot of work ahead, and plenty of ideas waiting to be explored.
My immediate focus will remain on improving the foundations of Llamatik, making it easier for developers to integrate local AI into their applications, and continuing the development of its agent capabilities.
What started as a personal experiment has become a project with a much broader vision.
This grant is an important milestone, but certainly not the destination.
I’m excited to see where Llamatik goes next.
Thank you for being part of the journey!