At INDEPENDENT, I’ve focused on LLM serving, inference pipelines, RAG backends, and LLMOps. I deployed and benchmarked vLLM inference serving with AWQ quantization, cutting latency 40% and raising throughput 2.5× versus TGI and Ollama baselines.
At Clario SRL, I rolled out Dynatrace across Kubernetes clusters and helped improve incident recovery time. Earlier, at Equifax Inc., I standardized multi-cloud provisioning with Terraform and CloudFormation, bringing environment build time from two weeks to under 45 minutes. I bring that platform and cloud experience to production AI infrastructure, where reliability and cost matter.
