Join Tether and Shape the Future of Digital FinanceAt Tether, we’re not just building products, we’re pioneering a global financial revolution. Our solutions enable seamless integration of reserve-backed tokens across blockchains, empowering businesses worldwide. Transparency and security are at the core of our operations.Our OfferingsTether Finance:
Featuring the trusted stablecoin
USDT
and digital asset tokenization services.Tether Power:
Eco-friendly energy solutions for Bitcoin mining.Tether Data:
AI and data sharing innovations like
KEET .Tether Education:
Digital learning for global growth.Tether Evolution:
Merging technology and human potential.Why Join Us?Work remotely with a global team, innovate in fintech, and be part of a fast-growing industry leader. Excellent English communication skills are essential.About the role:
As part of our AI model team, you will innovate in model serving and inference architectures, focusing on optimizing deployment strategies for AI systems across various hardware environments. Your work will involve developing scalable, efficient inference pipelines, setting performance benchmarks, and troubleshooting bottlenecks to enhance real-world AI performance.Responsibilities:
Design and deploy high-performance model serving architectures suitable for resource-limited environments.Establish and meet performance metrics such as latency, throughput, and memory usage.Conduct inference testing in controlled and live environments, tracking key performance indicators.Prepare datasets and simulation scenarios to evaluate model performance under operational conditions.Analyze and optimize computational efficiency, addressing bottlenecks related to processing and memory.Collaborate with teams to integrate optimized inference solutions into production pipelines, ensuring continuous improvement.Qualifications include a degree in Computer Science or related fields, with a focus on NLP, ML, or AI R&D. Proven expertise in kernel and inference optimization on mobile devices, experience with model serving frameworks, and a track record of improving inference performance are required.
#J-18808-Ljbffr