Lillup · 8 hours ago
Edge LLMOps Engineer Intern (On-Device LLM Release + iOS Integration) - (Remote, Unpaid)
Lillup builds on-device AI experiences for human capital and lifelong learning. They are seeking an Edge LLMOps Engineer Intern to work closely with the engineering team to help ship reliable edge LLM capabilities, focusing on model pack pipelines and iOS integration.
Responsibilities
Build and maintain model pack pipelines
Own an edge compatibility matrix: device/OS/runtime/model variant
Support iOS integration: stable inference contract, streaming/cancel behavior, lifecycle robustness, debugging OOM/latency issues
Maintain release discipline, reproducible builds
Qualification
Required
Strong CS/software fundamentals; comfortable debugging performance/memory
Experience with on-device/edge ML or LLM inference (projects are fine)
Familiar with at least one stack: llama.cpp/GGUF, Core ML, MLX, MLC, ONNX Runtime (Metal/Vulkan concepts a plus)
Clear communicator in a remote environment; able to coordinate tasks and document decisions
Preferred
iOS profiling (Instruments), Swift integration patterns, mobile performance tuning
Experience building eval harnesses or packaging/release pipelines