Building a low-latency multimodal agent pipeline running unquantized 70B+ open-weight models on AMD ROCm/vLLM for real-time telemetry inspection and automated audit trails. Looking for a systems engineer familiar with ROCm/vLLM kernel optimizations and a frontend developer skilled in high-frequency WebSocket dashboards.
.png&w=256&q=75)
.png&w=256&q=75)
.png&w=256&q=75)