player: Shubhransh Gupta
class: iOS Architect & On-Device AI / LLM Optimization Engineer
guild: Lenskart
specialization: Swift · CoreML · MLX · Metal · Quantization (INT4/AWQ) · LLM Inference
level: 5+ years XP · MakeMyTrip · Navi · Nuclei · Lenskart
status: ONLINE — Shipping edge AI & mobile architectures to millions
spawn: 🇮🇳 IndiaSelf-taught builder turned iOS & Edge AI architect. Specializing in high-performance iOS systems, on-device SLMs/LLMs, CoreML/Metal acceleration, and memory-constrained inference.
| Project / Org | Focus Area | Contribution Highlights |
|---|---|---|
| OfflineCV | Edge Computer Vision | On-device computer vision pipelines and offline edge processing. |
| Excalidraw | Collaborative Canvas | Core canvas tooling, performance optimizations, and UX enhancements. |
| MLX-Swift & Local AI | Apple Silicon & LLMs | Metal-accelerated inference, KV cache optimization, and Swift LLM tooling. |
| ShubhranshProxy | macOS Networking | Native macOS HTTP(S) debugging proxy with SSL decryption & HAR export. |
| Core Mobile & Systems | AI & Edge Inference | Architecture & Quality | DevOps & Tooling |
|---|---|---|---|
| Swift · SwiftUI · UIKit | CoreML · MLX-Swift · Metal | Clean Architecture · VIPER | Fastlane · GitHub Actions |
| Swift Concurrency · Combine | INT4 / AWQ / GGUF Quantization | MVVM · Dependency Injection | Docker · LLDB · Instruments |
| macOS AppKit · C / C++ | Tokenizers · KV-Cache · ONNX | XCTest · Snapshot Testing | Git · CI/CD Pipelines |
