Teams at Hyphen Connect Limited

Optimize and compress large language and vision models for on-device inference. Build model distillation and hardware-specific compilation pipelines, and benchmark performance across various NPU/GPU architectures to enable efficient edge deployment.

Design and architect multi-agent orchestration frameworks, build robust tool-calling pipelines and reasoning loops, and optimize latency and token usage for real-time agentic workflows. Apply expertise in multi-agent frameworks and agentic RAG to enable autonomous tool invocation and scalable workflow automation.

Design and orchestrate large-scale LLM pre-training across 1,000+ GPUs using PyTorch, DeepSpeed, or Megatron-LM. Optimize InfiniBand/RDMA networking and memory to avoid OOM, automate checkpointing and failure recovery for month-long runs, and manage SLURM or Kubernetes GPU clusters. Implement systems-level improvements using C++, CUDA, and Python.