Yaniv Hassidof
Robotics and AI PhD Candidate with 6+ years of experience.
Proven track record as lead developer on high profile products
and in leading integration of deep learning into high stakes robotic systems.
Passionate about leveraging my skills towards new technological advancements
Open for internship roles
Employment
2024 - Present: Teaching Assistant at Technion
- Teaching courses in Reinforcement Learning, Mobile Robots, and Computer Networks
- Supervised projects such as vision-impaired collision warning , LLM RAG teaching-assistant, and more
2020 - 2024: Navigation Algorithms Engineer at Rafael
- Taken full responsibility of the navigation stack for multiple high-profile Rafael systems, from localization and mapping to planning under uncertainty
- Single-handedly crafted POCs that proved superior to alternatives from across the company
- Led the department-wide adoption of Deep Learning, assisting dozens of projects and resulting in significant performance improvements in signal processing, state estimation, simulation, etc.
2018 - 2020: Software Developer at Self Reception Ltd
- Managed full-cycle project development for the Self Reception Hotel Kiosk
- Ensured customer success with continuous improvements and on-site support
Education
2024 - Present: MSc./PhD in Electrical Engineering, Technion
- Thesis: Developing a novel Diffusion Models-based Tree Search algorithm for safe, real-time, and agile robotic motion planning in complex, uncertain environments.
- Researching Physics-Informed Neural Networks (PINNs) to enhance the accuracy of real-world robotic simulations, bridging the gap between physics-based modeling and deep learning.
2019 - 2023: BSc. in Computer Engineering, Technion
- Specialized in Deep Learning and Quantum Computing
- Extracurricular courses from MIT and Stanford
- Featured on the Dean’s Excellence list twice
Projects
Drone Delivery Project
Developed vision and planning algorithms for multiple drone delivery systems.
Missile-Command
Evaluating Reinforcement Learning (RL) algorithms
within the context of a multi-agent competitive scenario. Implemented
in a mini strategy game as a testbed for multi-agent RL.
Assessed the efficacy, adaptability, and strategic depth
of various RL algorithms in decision making, cooperation, and competition in a simulated setting.
View Code on GitHub
Graph Neural Networks Rewiring using RL
Addressed the oversquashing phenomenon in GNNs by learning optimal edge modifications. View Code on GitHub