30+ days ago - req13423
Machine Learning and Data Engineer
Research & development
In a nutshell
Research & development
Within ASML the sector Development & Engineering is responsible for the development, specification and design of new ASML products. The Business Line Applications provides integrated solutions with computational, metrology and control technology. These solutions extend and improve the performance of lithography and patterning products for the semiconductor industry.
Do you enjoy solving algorithm design problems for semiconductor metrology industry, with demanding time, accuracy, and memory requirements ? Do you like to use your creativity, your in-depth knowledge of data engineering and machine learning principles, and your hands-on experience with practical problem solving, being part of a highly talented group of algorithm experts ?
Within D&E Applications, the group YieldStar Algorithms and Physical Modeling covers the development of models and methods required to infer physical model parameters from optical scatterometry data. Relevant new metrics, as well as new measurement functions, with optimum performance characteristics using the raw acquisitions are identified, designed and implemented. The group secures the ASML Competency of Applied Mathematics for Parameter Estimation.
-Act at the interface to colleague data science groups and SW groups in ASML, enabling cloud storage and cloud development architectures as environment for machine-learning and deep-learning techniques in our group.
-Drive for quality of the database architecture, and help direct colleagues in structuring compatible code architectures.
-Working as a team with similar-minded people, benefitting from each other’s specific competences.
-Propose solutions for statistically correct parameter inference, machine learning and optimization algorithms, and system calibrations, which improve semiconductor metrology and enable high-volume fab control solutions.
-Communicate crystal clearly on algorithm solutions, and architectures to stakeholders, without omitting the essentials.
-Design and realize fully functional proof-of-concept subsystems on the edge of system specifications, costs and project planning, thereby contributing directly to products for B2B customers world-wide.
-Review technical analyses from the team, and structure team contributions keeping the overview.
-Consolidate technical-team identity in communication with other departments.
-Contribute to technical product roadmaps and generate intellectual property protecting ASML products, while developing the best metrology solutions and a well-founded vision on semiconductor metrology.
Ph.D. in Computer Science, Electrical Engineering, Physics, or Applied Mathematics
-Experience in (setting up) cloud storage and cloud development architectures as environment for machine-learning and deep-learning techniques, and affinity with driving the quality of the database architecture
-Drive for structuring the scripting code architecture in the cluster, and be energized by helping colleagues in this
-Fluency in the language and standards of data structure design, and awareness of compatibility with other softwares
-Excellence in numerically stable modeling, code development, and in applying them to physical problems
-Expertise in design for manufacturability, stochastic programming and robust optimization
-Ability to explain complex architectures and algorithms in a crisp way, without omitting the essentials
-Sound understanding of the fundamentals such as linear algebra, probability theory, and (deep) learning methods
-Drive creative solutions -within the bigger picture- with the product and customer in mind
-Decisive and self-initiating in an ambiguous environment
-Team worker, and ability to influence without power
-Pragmatic approach and pro-active attitude, with result focus and a ‘can do’ spirit
Context of the position
This position is available within the group YieldStar Algorithms and Physical Modeling thatcovers the development of models and methods required to infer physical model parameters from optical scatterometry data.
Keywords: parameter inference, (non-)convex and robust optimization, deep learning, (un)supervised and reinforcement learning, neural network, inverse problems, surrogate modeling, classification, regression, stochastic programming, information theory.