AI Inference Engineer – QVAC

This is a fully remote position, open to applicants anywhere in the world.

📋 Description

• Take ownership of the inference backbone that supports QVAC’s local AI infrastructure.

• Develop the C++ systems layer to ensure rapid, dependable, and consistent model execution on user hardware.

• Enhance startup performance, manage memory usage, optimize throughput, reduce latency, and maintain long-session stability.

• Define and advance fundamental abstractions for inference capabilities.

• Implement machine learning models on edge devices utilizing llama.cpp and ggml.

• Adapt and improve inference engines for efficient operation across various hardware platforms.

• Collaborate with researchers on coding, training, and transitioning models from research phases to production environments.

• Integrate AI functionalities into current products.

• Ensure the stability, optimization, and readiness of the inference layer for stack integration.

⛳️ Requirements

• Proficient programming skills in C++.

• Extensive experience with Llama.cpp and ggml inference engines.

• Familiarity with at least one GPU framework: CUDA, Vulkan, Metal, or OpenCL.

• Strong understanding of deep learning principles and model architectures.

• Experience with transformers, LLMs, and diffusion models.

• Proven ability to quickly learn and adapt to new technologies and methods.

• A degree in Computer Science, AI, Machine Learning, or a relevant field.

• A solid history in AI research and development.

• Knowledge of training or fine-tuning LLMs is advantageous.

• Experience in deploying models is a plus.

• Research background in novel model architectures is a bonus.

• Familiarity with distributed systems is an advantage.

• Experience with JavaScript is a bonus.

🏝️ Benefits

• Work remotely from any location around the globe.

• Collaborate with a diverse global remote team.

• Engage in private, on-device AI experiences and peer-to-peer AI products.

• Contribute to innovative projects in fintech, blockchain, AI, and digital asset domains.