ProfileSenior iOS/ macOS engineer with 15+ years building and shipping Apple-platform software across fintech, security-sensitive and performance-critical environments. Strong in modern Swift, Swift 6/ strict concurrency, SwiftUI, architecture, CI, debugging and production releases. Currently focused on AI-native Apple-platform and agentic systems engineering: combining senior mobile architecture with OpenAI agent workflows, local/on-device inference and privacyconscious LLM-assisted delivery. Comfortable across both major implementation paths: hosted Agent Builder + ChatKit workflows for product-facing agent chat experiences, and local/codefirst OpenAI Agents SDK implementations when the application needs to own orchestration, tools, approvals, state, storage, deployment and runtime behavior. Practical with Codex, Claude, CLI agents, repo-local instructions, skills, plugins, MCP tools, worktrees, review loops and verification commands for implementation, refactoring, migration planning, documentation, debugging and code review. Hands-on ML background including custom neural networks and classifiers, TensorFlow 2 coursework/certification, and local inference experiments using Core ML, ML Kit, and Appleplatform runtime integration. Interested in AI systems that are not just demos, but maintainable, testable, privacy-aware features embedded into real applications. Systems thinker with strong pattern recognition, enabling fast end-to-end mental models, early detection of hidden coupling and design for real operational constraints.
CORE STRENGTHS- Swift 6 & concurrency correctness: async/await, actors, Sendable, strict concurrency checks, TaskGroups, cancellation; race condition debugging
- SwiftUI stack: SwiftUI, Observation (@Observable), SwiftData; UIKit/AppKit interoperability
- Charts & visualization: Swift Charts, custom rendering where needed; performance tuning (Instruments, signposts), Metal when beneficial
- Synchronization: Synchronization framework (Mutex) for safe shared state where actors aren?t the right trade?off
- Architecture: MVVM(+C) + Coordinator, Clean Architecture, Clean Swift (VIP), VIPER, MV (Model?View) SwiftUI (view?model?less / Observation?driven)
- Systems thinking (pattern?first cognitive style): tends to process information differently, surfacing hidden coupling, edge cases, and second?order effects early; translates that into pragmatic architecture and clean interfaces.
- Quality & delivery: Swift Testing / XCTest, UI tests, snapshot tests, integration tests, CI/CD, code review, SDLC/SAFe environments
- Security mindset: auth/session design, secure config handling, reverse?engineering awareness
- AI-native/ agentic development
- OpenAI agent stack: hosted Agent Builder + ChatKit workflows, local/code-first OpenAI Agents SDK implementations, Responses API concepts, tool calling, MCP/connectors, guardrails, handoffs, traces, evals, and human review
- Agentic workflows: Codex, Claude, CLI agents, repo-local instruction files, skills, plugins, subagent/parallel exploration, and task-specific orchestration
- LLM workflow design: context shaping, prompt/instruction design, multi-step task decomposition, tool routing, workflow state, verification loops and human-reviewed output
- Software delivery with agents: using LLM agents for implementation, refactoring, test creation, documentation, migration planning, debugging, and review support
- On-device AI: Core ML, ML Kit, local inference architecture, Apple-platform model integration
- ML foundations: custom neural networks/classifiers, TensorFlow 2, model training/evaluation basics
- Privacy-first practice: no secrets/client-data leakage, minimal/redacted context, policycompliant usage and reviewed/tested output before integration
Skills- UI:
- Modern Swift:
- Swift 6, Structured Concurrency (async/await, actors, Sendable), Synchronization (Mutex), GCD (legacy/interop)
- Frameworks:
- Observation, Swift Charts, SwiftData, Core Data, Combine/RxSwift (interop), Swift?NIO (when applicable), Metal
- AI Agentic development:
- OpenAI Agents SDK, Agent Builder, ChatKit, hosted agent workflows, local/code-first agent orchestration, Codex, Claude, CLI agents, skills/plugins, MCP, LLM workflows, prompt/instruction design, verification loops
- Codex workflow control:
- Plan mode, Goal mode, worktrees, repo-local instructions, custom skills/plugins, MCP tools, review workflows, subagent/parallel task decomposition
- AI/ ML:
- Core ML, ML Kit, local/on-device inference, TensorFlow 2, neural networks, classifiers
- Testing:
- Swift Testing, XCTest, UI Testing, snapshot tests, integration tests
- Architecture:
- MVVM(+C) + Coordinator, Clean Architecture, Clean Swift (VIP), VIPER, MV (Model?View) SwiftUI (view?model?less/ Observation?driven)
- Tooling:
- Xcode, Instruments, Swift Package Manager, Git, Charles Proxy, Wireshark, Figma
- CI/DevOps:
- ?GitHub/GitLab, CI pipelines, secure environment config