Teams at Tether.io

Design and optimize model-serving architectures and inference pipelines for high-throughput, low-latency AI performance on mobile, edge, and GPU-cluster environments. Develop custom GPU kernels and Metal shaders, benchmark production systems, diagnose computational and memory bottlenecks, and implement techniques including parallelism, quantization, pruning, FlashAttention, KV caching, and speculative decoding. Collaborate cross-functionally to integrate scalable inference solutions and continuously improve performance.

Coordinates event-related creative requests from intake through delivery, briefing designers, managing timelines, reviewing print and digital assets, maintaining documentation, coordinating vendors, and identifying production or logistical risks. The role ensures brand, file, print, and platform specifications are met while improving event creative workflows and keeping cross-functional stakeholders aligned.

Lead architecture, delivery, and engineering management for a bare-metal GPU infrastructure platform. Responsibilities include operating Slurm and Kubernetes, GPU fleet lifecycle, high-performance networking, managed inference, observability, reliability, partner coordination, capacity planning, hiring, and hands-on technical leadership across a distributed team.