Experience

Giulio Romualdi, Ph.D.

Professional Summary

I turn advanced humanoid robotics research into working robotic products. As Robotics Technical Lead and Product Owner at Generative Bionics, I lead the development and delivery of locomotion and whole-body control technologies for humanoid robots — owning the technical roadmap, priorities, backlog and delivery criteria, driving technical decision-making, and mentoring engineers and researchers. Alongside my leadership responsibilities I stay hands-on across the full locomotion stack, from reinforcement-learning policy design and motion generation to simulation, deployment, experimental testing and on-robot tuning. My background combines deep robotics research — nonlinear and model predictive control, optimization, whole-body control and learning — with delivery on real humanoid platforms including iCub, iCub3, ergoCub and GENE.01.

Experience

• 2026 – Present
  🚀 Robotics Technical Lead & Product Owner — Generative Bionics
  Technically lead the development and delivery of locomotion and whole-body control technologies for humanoid robots, and act as Product Owner of the locomotion team. In this role I:
  • translate technical and product objectives into roadmap, priorities, backlog, milestones, Definition of Done and acceptance criteria;
  • lead technical execution and drive technical decision-making, evaluating alternative approaches with the team and making the final implementation call when needed;
  • mentor engineers and researchers, provide technical feedback and help resolve technical disagreements;
  • participate in and organize technical hiring interviews and contribute to candidate evaluation;
  • report progress and results to senior leadership and work cross-functionally with other engineering teams, including providing simulation data and technical input to the mechanical team for future humanoid platforms;
  • remain technically hands-on across RL policy design, motion generation, simulation, sim-to-sim validation, deployment and on-robot tuning.
  🏆 Led the locomotion delivery for GENE.01, contributing to its public live demonstration at AMD Advancing AI 2026.
• September 2022 – January 2026
  🔬 Postdoctoral Researcher — Istituto Italiano di Tecnologia, Artificial and Mechanical Intelligence
  Humanoid locomotion, whole-body control and locomotion-architecture development. Helped coordinate researchers working on robust, reactive and dynamic locomotion, and mentored PhD students and researchers on research ideas, experiment design, code review, software development and scientific writing. Contributed to locomotion architectures for iCub, iCub3 and ergoCub. Served as IIT technical point of contact for the ergoCub project (IIT–INAIL collaboration), organizing monthly meetings with INAIL, presenting the state of work and coordinating technical communication about IIT's activities.
  🏆 ergoCub robot featured walking on national Italian television.
• 2018
  🧪 Research Fellow — Istituto Italiano di Tecnologia
  Humanoid locomotion research bridging the M.Sc. thesis and the start of the PhD.
• 2017 – 2018
  🎓 Guest Student — Istituto Italiano di Tecnologia
  Dynamic humanoid locomotion during the M.Sc. thesis.
• University of Pisa · during studies
  📚 Part-time Student Worker – University Library Services
  Worked part-time in university library services while completing engineering studies, supporting book reservations and related student services.

Education

• 2018 – 2022
  🎓 PhD in Mechatronics, Robotics and Automation Engineering — University of Genoa / Istituto Italiano di Tecnologia
  Thesis: Online Control of Humanoid Robot Locomotion. Focus on humanoid locomotion, nonlinear control, model predictive and optimal control, optimization and whole-body robotics, with experimental validation on iCub and iCub3 and within the ANA Avatar XPRIZE activities (locomotion + teleoperation integration).
• October 2021 – December 2021
  🌍 Visiting PhD Student — LAAS-CNRS, Toulouse
  Whole-body control and state estimation for humanoid robots with structural/link flexibility (TALOS). Led to the first-author paper Whole-Body Control and Estimation of Humanoid Robots with Link Flexibility (IEEE-RAS Humanoids 2022).
• 2015 – 2018
  📘 M.Sc. in Robotics and Automation Engineering — University of Pisa · 110/110 cum Laude
  Thesis: Capture-Point Based Controllers for Robot Bipedal Locomotion: Analysis and Implementation on the iCub Platform.
• 2011 – 2014
  📗 B.Sc. in Biomedical Engineering — University of Pisa

Professional Service & Recognition

• 🗂 Organizing committee — Designing Interactive Humanoids: Learning Tasks through Interaction with Humans, IEEE-RAS Humanoids 2024
• 🏅 Finalist – Best Oral Presentation Award, IEEE-RAS Humanoids 2024 — Online DNN-driven Nonlinear MPC for Stylistic Humanoid Robot Walking with Step Adjustment (details)
• 🤖 ANA Avatar XPRIZE Finalist – iCub Team — contributed the integration of humanoid locomotion with the teleoperation system used in the iCub3 avatar activities (team achievement)
• 🥈 Runner-up Best Poster Award — ADHERENT, NAVER LABS Europe AI for Robotics Workshop, 2021 (co-author, team recognition)
• 🏅 Finalist – Best Interactive Presentation, I-RIM 2019 — Online Trajectory Generation for Step Recovery in Bipedal Robot Locomotion

Selected Publications

• 2024: Online DNN-driven Nonlinear MPC for Stylistic Humanoid Robot Walking with Step Adjustment
  📚 Conference: IEEE International Conference on Humanoid Robotics (Humanoids)
  Paper, Website, Project page
• 2024: [Workshop] Designing Interactive Humanoids: Learning Tasks through Interaction with Humans
  🗂 Conference: IEEE International Conference on Humanoid Robotics (Humanoids)
  Workshop Website
• 2024: iCub3 Avatar System: Enabling Remote Fully Immersive Embodiment of Humanoid Robots
  📰 Journal: Science Robotics
  Paper
• 2022: Whole-Body Control and Estimation of Humanoid Robots with Link Flexibility
  📖 Conference: IEEE International Conference on Humanoid Robotics (Humanoids)
  Paper
• 2022: Dynamic Complementarity Conditions and Whole-Body Trajectory Optimization for Humanoid Robot Locomotion
  📖 Journal: IEEE Transactions on Robotics (T-RO)
  Paper
• 2022: Online Non-linear Centroidal MPC for Humanoid Robot Locomotion with Step Adjustment
  📚 Conference: IEEE International Conference on Robotics and Automation (ICRA)
  Paper, GitHub
• 2022: ADHERENT: Learning Human-like Trajectory Generators for Whole-body Control of Humanoid Robots
  📰 Journal: IEEE Robotics and Automation Letters (RA-L)
  Paper
• 2021: Modeling of Visco-Elastic Environments for Humanoid Robot Motion Control
  📚 Journal: IEEE Robotics and Automation Letters (RA-L)
  Paper, GitHub
• 2019: A Benchmarking of DCM Based Architectures for Position, Velocity and Torque Controlled Humanoid Robots
  📖 Journal: International Journal of Humanoid Robotics
  Paper, GitHub

📚 For the complete list of publications, visit my Google Scholar profile.

Skills & Competencies

• Leadership & Product: 🧭 Technical leadership, product ownership, roadmap & backlog management, Definition of Done and acceptance criteria, technical decision-making, mentoring, technical hiring, cross-functional collaboration, stakeholder reporting
• Robotics: 🤖 Humanoid locomotion, whole-body control, model predictive & nonlinear control, optimal control, optimization, state estimation, reinforcement learning for locomotion, simulation and sim-to-real deployment
• Software: 💻 C++, Python, MATLAB, PyTorch, Isaac Lab, Eigen3, iDynTree, osqp, IPOPT, CasADi, Qt, YARP, ROS2, Git, CMake

Professional Qualification

• 🎖️ Qualification to practise as an Engineer — Section A, Industrial Engineering Sector, State Examination at the University of Pisa (Second Session 2020, published February 2021).

Languages

• 🇮🇹 Italian — Native / bilingual proficiency
• 🇬🇧 English — Full professional proficiency
• 🇯🇵 Japanese — Elementary proficiency

Open-Source Software Maintainer

• Bipedal Locomotion Framework — suite of libraries for whole-body control and locomotion of humanoid robots: GitHub
• osqp-eigen — C++ Eigen wrapper for the OSQP quadratic-programming solver: GitHub
• walking-controllers — DCM-based walking controllers for humanoid robots: GitHub
• AMP-RSL-RL — Adversarial Motion Priors for reinforcement-learning-based locomotion: GitHub