At Sthenos AI, part of EFA group, I owned RAG products from architecture through implementation, deployment, and optimization. I also deployed distributed LLM inference with Ray Serve on Kubernetes and developed predictive maintenance models using IoT sensor data.
At Engys, I built PyTorch models and ML pipelines for automotive applications in the Upscale project, which aimed to reduce EV development time. I also developed performance-critical C++ simulation software and parallel code for HPC systems; earlier, I worked on CFD solvers and computational methods at City St George’s, University of London and other research organizations.
