Back to Projects
AIPublicFeatured

DevIO

AI-powered developer tool that provides a secure mobile interface for locally hosted LLMs (Ollama), with a focus on data privacy and offline functionality.

DevIO turns a phone into a secure control interface for local LLM servers. It is designed for developers who want fast AI workflows without sending sensitive data to third-party cloud services.

FlutterDartAILLMsBLoC

Role

Product Engineer

Timeline

Ongoing

Delivery

Structured implementation across design, engineering, and release phases.

Challenge

  • Build a clean mobile UX for complex LLM interactions.
  • Keep model communication secure across local networks.
  • Support practical offline and low-connectivity workflows.

Solution

  • Implemented a Flutter app with a modular architecture and predictable state management.
  • Designed a local-first connection flow to servers like Ollama.
  • Optimized interactions for quick prompts, history, and session continuity.

Architecture

  • Flutter client with feature-based modules.
  • Local LLM backend compatibility (Ollama-first design).
  • Reusable service layer for model requests and response parsing.

Outcomes

  • Shipped to the App Store with production-level UX.
  • Enabled practical AI usage with stronger privacy guarantees.
  • Established a base for additional AI workflows in mobile.

Next Steps

  • Expand multi-model routing options.
  • Add richer context management for longer sessions.
  • Introduce advanced developer tooling around prompt templates.

More Projects