Senior AI Inference Engineer (llama.cpp)

TL;DR

Senior AI Inference Engineer (llama.cpp): Developing and optimizing the C++ layer for local AI, specifically porting and enhancing inference engines like llama.cpp and ONNX to run efficiently on edge devices. Focus on runtime optimization for faster loading, leaner execution, and superior performance across diverse hardware, enabling private and fast on-device AI without cloud infrastructure reliance.

Location: Fully remote, worldwide

Company

hirify.global is pioneering a global financial revolution by offering cutting-edge solutions for integrating reserve-backed tokens across blockchains, featuring the world’s most trusted stablecoin, USDT, and expanding into energy, AI, and education.

What you will do

• Develop the C++ layer for local AI, enhancing inference engines like llama.cpp and ONNX for edge devices.
• Optimize model runtime for faster loading, leaner execution, and performance across different hardware.
• Ensure the inference layer is stable, optimized, and ready for integration into products.
• Collaborate with researchers to transition machine learning models from research to production environments.
• Integrate advanced AI features into existing products.

Requirements

• Excellent programming skills in C++.
• Strong experience with Llama.cpp and ggml inference engines.
• Good understanding of deep learning concepts and model architectures.
• Experience with transformers and LLMs.
• Proven ability to rapidly assimilate new technologies.
• Degree in Computer Science, AI, Machine Learning, or related field with AI R&D track record.
• Excellent English communication skills.

Culture & Benefits

• Global remote team, fostering collaboration from anywhere in the world.
• Opportunity to innovate in the fintech space with a market leader.
• Work with cutting-edge technology to build a global financial revolution.
• Be part of a lean, fast-growing company setting new industry standards.

Hiring process

• Application requires detailed experience descriptions (e.g., C++, llama.cpp, edge device deployment).
• Candidates must provide an expected annual salary in USD.
• Use of AI tools for application answers is prohibited and may lead to disqualification.