Gunnar Hostetler

I have 3 iOS apps on the App Store. I built all of them with AI. I'm currently working on my fourth — a RAG engine that runs completely on-device. I build because I genuinely can't turn this thing off.

View My Work Get In Touch

About Me

I wanted an iOS client for OpenAI's Assistants API. OpenAI never made one. So I figured it out and built OpenAssistant.

Then I wanted to do RAG with Pinecone on my phone. So I built OpenCone.

Then Assistants got deprecated and I needed the new Responses API with all the tools. So I built OpenResponses. It passed App Review on first submission.

Now I'm working on OpenIntelligence — hybrid search, on-device embeddings, Apple Intelligence routing, no cloud calls unless you explicitly consent. It's the most complex thing I've built and I'm genuinely fascinated by it. It runs offline.

I work full-time as a medical device specialist at the VA in Palo Alto. I build at night and on weekends because this is what my brain wants to do. I've figured things out this far, and I trust I'll keep figuring them out as I go.

Gunnar Hostetler

How I Build

I see a problem, I figure out how it should work, and I don't stop until it's on the App Store. The code comes from models—the vision is mine.

OpenResponses commits
OpenIntelligence commits
OpenCone commits
OpenAssistant commits

What I Own

Product Vision System Architecture RAG Pipelines API Integration Privacy Design Prompt Engineering

The Stack

Swift / SwiftUI OpenAI APIs Anthropic APIs Pinecone Xcode Git / GitHub VS Code + Copilot

Healthcare Ops

Clinical Operations HIPAA Compliance Medical Devices Surgical Workflows

The Open- Series

OpenResponses iOS App Icon

OpenResponses

Stars Last Commit

The successor to my Assistants-era client, rebuilt as an iOS OpenAI Responses API playground with AppContainer dependency injection, SwiftUI, and MVVM handling Responses streaming events.

  • Model routing: Configurable routes optimized for GPT-5.x (including gpt-5.2 / gpt-5.2-pro / gpt-5.1) plus GPT-4.1/4o mini and computer-use-preview, with cancellable runs, reasoning traces, and per-conversation parameters.
  • Tool coverage: Supports Computer Use preview, Code Interpreter (sandboxed Python + container file retrieval; artifact preview for images and common text/data formats), File Search, Web Search, Image Generation, and custom function schemas.
  • MCP integrations: Includes starter Model Context Protocol connector configs (e.g., Notion, Gmail, SharePoint, Dropbox) plus remote server templates, all keychain-backed with approval gates.
  • Streaming telemetry: OpenAIService streams token counts, tool execution cards, and status chips across iPhone and iPad.
SwiftUI Combine OpenAI Responses API GPT-5.x Model Context Protocol Computer Use MCP Python Sandbox File Search Web Search Async/Await
Docs GitHub App Store
OpenIntelligence

OpenIntelligence

Stars Last Commit

An on-device RAG engine (RAGMLCore) built for iOS with SemanticChunker, NLEmbedding, and hybrid retrieval fused through RRF + MMR, all wrapped in a consent-aware SwiftUI interface.

  • Document ingestion: PDFKit + Vision OCR feed SemanticChunker (400-word target, 75-word overlap) before on-device embeddings via NLEmbedding.
  • Hybrid retrieval: Reciprocal Rank Fusion combines cosine similarity and BM25 lexical scores, then MMR diversification assembles citations.
  • Service graph: ContainerService, RAGService, HybridSearchService, and PersistentVectorDatabase coordinate ingestion, retrieval, and storage per isolated KnowledgeContainer.
  • Privacy + consent: Cloud calls are blocked until Telemetry-backed consent is granted, with per-provider revocation and on-device observability.
On-Device RAG Consent Telemetry On-Device Analysis SwiftUI Swift 6 RAG On-Device Embeddings LLM Integration Local Storage CloudKit Core ML Hybrid Search Async/Await
Docs GitHub
OpenCone iOS App

OpenCone

Stars Last Commit

A shipped iOS knowledge base that ingests local documents, builds Pinecone serverless indexes, and streams OpenAI Responses answers with inline citations.

  • Processing pipeline: DocumentPicker + security-scoped bookmarks feed PDFKit/Vision OCR, MIME-aware chunking, and OpenAI embeddings before Pinecone upserts.
  • Search flow: User queries are embedded, top-k Pinecone matches stream into Responses model answers with inline citations.
  • Operational telemetry: DocumentsView + DocumentDetailsView expose timing charts, token counts, and per-step logs for every ingestion.
  • SwiftUI system: MVVM feature modules (Documents, Search, Logs, Settings) share the OCDesignSystem themes, with Keychain-backed credential storage (including legacy UserDefaults migration) and persistent UI preferences.
SwiftUI OpenAI Responses API Pinecone API PDFKit Vision OCR RAG Async/Await
Docs App Store GitHub
OpenAssistant iOS App

OpenAssistant

Stars Last Commit

The original Assistants API (v2) client that managed assistants, vector stores, and tool stacks before the Responses API shipped; now maintained as a legacy reference.

  • Assistant lifecycle: CRUD workflows for models, instructions, tool toggles, and deletion built atop async/await + Combine.
  • Vector store tooling: Complete CRUD plus file uploads and store-to-assistant linking.
  • File + tool handling: Multipart uploads feed Code Interpreter and File Search with Keychain-backed credential storage.
  • Threaded chat: Markdown-rendered chat UI mirrors OpenAI threads with streaming SSE updates and persistent history.
SwiftUI OpenAI Assistants V2 API (now deprecated) Code Interpreter File Search
Docs App Store GitHub

Professional Experience

Independent Builder

Building the Open- Series

Nov 2023 – Present

Leveraging LLMs to build tools. Since my first commit in late 2023, I've used Foundation Models to generate, refine, and ship complex native iOS applications. Driven by a passion for exploration, I orchestrate systems involving RAG, Agents, and on-device intelligence to close the gaps I see in existing software.

  • OpenResponses: Production successor to my Assistants client with Responses API streaming, Computer Use, MCP connectors, and core tool coverage.
  • OpenIntelligence: Privacy-first RAGMLCore engine running on Apple silicon with on-device embeddings, hybrid search, and consent gating.
  • OpenCone: App Store RAG pipeline that pairs Pinecone serverless indexes with OpenAI Responses answers and detailed ingestion analytics.
  • OpenAssistant: Legacy SwiftUI client covering Assistants, Threads, Runs, and vector stores before the Responses API launched.

OnSite Specialist

Stryker | VA Palo Alto Health Care System

Aug 2022 – Present

Sole technical specialist supporting Stanford surgical teams with complete autonomy. Bridging the gap between complex medical technology and clinical workflows in high-pressure surgical environments.

  • Intraoperative Support: Provide real-time technical support for General, Urology, ENT, and Thoracic surgeries, serving as the subject matter expert for advanced medical devices.
  • Operations & Procurement: Manage end-to-end procurement and inventory for surgical supplies, forecasting needs based on surgical schedules to prevent case delays.
  • Technical Liaison: Act as the primary bridge between surgeons, nursing staff, and biomedical engineering, translating technical constraints into clinical solutions.
  • Compliance & Safety: Maintain strict HIPAA compliance and government clearance (PIV) while managing sensitive case data and equipment protocols.

Get In Touch

Let's Connect

I'm always interested in new opportunities and collaborations, especially in iOS development and AI integration. Feel free to reach out if you'd like to discuss a project, explore potential partnerships, or just say hello!

Gunnarguy@me.com
iOS Engineer | AI Specialist

© 2025 Gunnar Hostetler. All rights reserved.