Résumé / October 2026
George Sterpu, PhD
Senior AI Engineer · Bray, Ireland · EU citizen
Profile
Senior AI engineer with a PhD from Trinity College Dublin and 9+ years in applied machine learning research and industrial R&D. Experience in speech, multimodal learning, and building automation.
Experience across problem formulation, architecture, experimentation, evaluation, optimisation, deployment, and technical collaboration. Areas include GNNs, Transformers, multimodal learning, speech AI, diffusion models, and Edge AI.
Experience
Senior Advanced AI Engineer · Honeywell
Jun 2025 — present
Graph neural networks for building topology, deployed to a governed endpoint on Azure Databricks. Also working on site onboarding and HVAC fault detection and diagnosis.
Product Manager & Support Specialist · Herbst Software
Aug 2024 — Jun 2025
Customer requirements, product management, software development, and DevOps for ERP-connected web software.
Senior R&D Engineer · Xperi
Feb 2021 — Apr 2024
Developed an end-to-end speech recognition engine for Ergo, running in real time with one-second latency at 100 mW. Also developed speech synthesis engines and explored model compression.
Doctoral Researcher · Trinity College Dublin
Sep 2016 — Dec 2020
Multimodal learning, audio-visual speech recognition, AV Align, Taris, and open-source research tools.
Earlier experience
Lab Assistant, Trinity College Dublin (2017–2019): deep learning tutorials and student support. Research Intern, INRIA Grenoble (2015): real-time sound source localisation. Software Engineer, CODECart Alliance (2013–2014): speech compression algorithms for the NXP Coolflux DSP.
Education
PhD · Trinity College Dublin
2016–2021. Audio-visual speech recognition, Electronic & Electrical Engineering and the ADAPT Centre.
Master 2 · University Grenoble Alpes
2014–2015. Electrical Engineering and Control Systems (MiSCIT), research track.
Bachelor’s · Politehnica University of Bucharest
2009–2013. Automatic Control and Computer Science, Systems Engineering.
Research
Seven first-author publications, including two journal papers and five conference papers. Open-source Python projects in multimodal learning and speech recognition.