Machine Learning Engineer

Remote

May 19, 2025

Full Time

Application ends: August 19, 2025

Job Description

We’re looking for a Machine Learning Engineer with a deep focus on optimizing models for real-world performance and deploying them on resource-constrained environments. You won’t just train models—you’ll figure out how to squeeze every drop of performance out of them for low-latency inference, quantization, pruning, and edge deployment scenarios. This role is ideal for someone who understands both theory and systems, and can bridge the gap between cutting-edge research and stable, scalable production.

What You’ll Do

Optimize model architectures for inference efficiency (e.g., using quantization-aware training, structured pruning, tensor decomposition techniques).
Build pipelines for automatic benchmarking of models across CPU, GPU, and mobile chipsets (e.g., Qualcomm, Apple Neural Engine).
Develop conversion workflows using tools like ONNX, TensorRT, Core ML, and TFLite for cross-platform compatibility.
Collaborate closely with embedded systems engineers and data pipeline teams to make ML models production-ready.
Profile inference bottlenecks using tools like Nsight Systems, PyTorch Profiler, and Perfetto, then implement strategies to minimize compute and memory overhead.
Evaluate and implement distillation strategies for large language models and vision models under tight performance constraints.
Stay up-to-date with hardware-aware neural architecture search (NAS) techniques and apply them when appropriate.

What You’ll Need

Required

Demonstrated experience in model compression, quantization, or low-rank approximations (show us your projects or papers).
Strong coding skills in Python and C++, with an emphasis on writing production-grade inference code.
Familiarity with TensorFlow Lite, ONNX Runtime, TensorRT, or other edge deployment frameworks.
Solid grasp of computer architecture concepts (cache hierarchy, vectorization, parallel execution).
Experience profiling ML workloads on devices with limited compute (e.g., Raspberry Pi, Jetson Nano, Android/iOS phones).

Preferred

Prior experience shipping ML models on consumer hardware (e.g., cameras, wearables, phones).
Contributions to open-source libraries or tools in the ML inference ecosystem.
Background in training custom quantization-aware or sparsity-aware models.
Working knowledge of compiler-based optimization for ML (e.g., XLA, TVM, Glow, MLIR).

Are you interested in this position?

Apply by clicking on the “Apply Now” button below!