Location: Eindhoven, the Netherlands (on-site, High Tech Campus) | Type: Full-time | Level: Medior, about 3 years of relevant experience
About us
AIM Solutions is an early-stage deep-tech startup in Eindhoven building inspection and metrology equipment for the semiconductor industry. We are in active build phase, small, and growing.
The role
You turn raw measurements into reconstructed images and explain what the images tell us. You work between the optics lab and the software team, and you become our in-house expert on this workflow and upcoming applications of our machine for the semiconductor manufacturing market.
You will:
Run the path from measurement to result: decide what to measure, run scans on the lab setup, configure and run reconstructions in our Python/JAX pipeline, and judge the quality of the output
Turn test cases from industry partners into measurement and reconstruction recipes, and automate them into a single-button-press workflow
Analyze results rigorously and investigate when a reconstruction does not look right: artifacts, convergence, noise, alignment
Drive the experiment loop with the optics and hardware engineers, and feed findings back into the algorithms and the tool design
Your profile
About 3 years of hands-on experience in a technical role (industry, PhD, or equivalent) with imaging, optics, or scientific data analysis
Practical experience with phase retrieval or coherent imaging, for example ptychography, CDI, holography, or Fourier-optics-based reconstruction. You have run reconstructions and interpreted the results
Strong Python, with NumPy and SciPy. You write reproducible analysis code and maintain a pipeline from data ingest to results
An experimental mindset: you design clean numerical experiments and know what to change to improve a result
You work well with optics and hardware engineers and turn vague questions into repeatable tests
Full professional proficiency in English
Bonus:
Hands-on experience with computational imaging and iterative reconstructions
PyTorch or JAX, GPU computing
Numerical optimization (gradient-based methods)
Machine learning on physics-based or scientific data
HDF5 and large datasets, Linux
We care more about rigor and learning speed than about a perfect match. If you are earlier in your career but have worked with reconstruction data, apply anyway.
What we offer
Ownership of the measurement-to-results workflow of a tool that does not exist anywhere else yet
Small team, short feedback loops, real impact on the product
Competitive salary for the Dutch market