How Tiles works
Building the future of software personalization with private memory networks.
Philosophy
We’re building open-source technology for local-first models to make personalized software experiences possible without compromising on accessibility or privacy. We believe identity and memory are two sides of the same coin, and Tiles makes that coin yours: your user-agent. Our first product is an on device memory management solution for privacy conscious users, paired with an SDK that enables developers to securely access a user’s memory and customize agent experiences.
We’re also working on the Personal Plus product, that adds support for multi-device memory sync and training with friends and family. It’s a one-time purchase with a lifetime license for each generation. Future generations will require a new license key to unlock added features.
The project has two values: simplicity and pragmatism. These values will help people understand what’s going on, helping us to go further and faster.
The project is defined by four design choices¹ that support and reinforce one another:
Device-anchored identity with keyless ops: Clients must be provisioned through the device key chain and cannot access the registry by identity alone². Keyless operations are enabled only after an identity is verified and linked to the device key, allowing third-party agent access under a user-defined policy³.
Immutable model builds: Every build is version-locked and reproducible, ensuring consistency and reliability across updates and platforms.
Content-hashed model layers: Models are stored and referenced by cryptographic hashes of their layers, guaranteeing integrity and enabling efficient deduplication and sharing.
Verifiable transparency and attestations: Every signing and build event is recorded in an append-only transparency log, producing cryptographic attestations that can be independently verified. This ensures accountability, prevents hidden modifications, and provides an auditable history of model provenance across devices and registries.
Roadmap
Our software stack includes a macOS app and a Modelfile⁴ based SDK. The Tiles app functions as a transparent, protocol-based proxy between the user and AI agents, leveraging a fine-tuned model to manage context and memories locally on-device, with linked references for graph visualization in tools like Obsidian. Logseq and iCloud support are coming soon. Our first-generation prototypes will launch as a CLI, with later versions introducing a patched build of OpenWebUI as the interface, and integrations for Iroh for multi-device P2P connectivity. While Tilekit, our Modelfile based SDK, lets developers to customize local models and agent experiences within Tiles. Our goal is to evolve Modelfile in collaboration with the community and establish it as the standard for model customization.
We automatically select and run the most suitable model for your device. We use fine tuned open models, with a continual learning memory setup, starting with qwen3:4b through Dria’s mem-agent model, trained for memory augmentation, semantic routing, tool calling, and other capabilities that enhance local-first personal applications. Over time, we ship out of the box with Obsidian support for storing memory and skills folders. Logseq and iCloud support are coming soon, while adding even smaller models such as qwen-1b and gemma3n:e2b.
We are building a high-end general system based on the hypothesis that abundant compute will continue to become more accessible on local consumer devices. We’re starting with Apple platforms, as their unified memory architecture and strong platform security make them well-suited for running private local models.
As part of our research on private software personalization infrastructure, we are investigating sparse memory finetuning, text diffusion models, Trusted Execution Environments (TEEs), Per-Layer Embeddings (PLE) with offloading to flash storage. In the next phase of our development, we will focus on our identity system model built on public key cryptography and verifiable attestations.
We are seeking design partners for training workloads that align with our goal of ensuring a verifiable privacy perimeter. If you’re interested, please reach out to us at feynon@tiles.run.