Python Developer Medical Imaging (3-D Reconstruction - Measurement)

Amplifier AI is building a surgical planning
platform powered by our own imaging,
segmentation and 3-D measurement engine.
We are a small team of engineers, radiologists
and data scientists from Ukraine and across
Europe. You would join the team developing the
internal Python engine behind 3-D
reconstruction and measurement for CT-based
workflows.

Read this before you apply

We hire for reasoning about 3-D space, not
for a title. The code itself is ordinary clean
Python — the hard part is everything the
geometry touches:
Voxel space vs. physical space, and every
transform between them
Origin, spacing and direction — consistent
across scanners, series and vendors
Real hospital data: incomplete series, odd
orientations, inconsistent metadata
Geometric edge cases — surfaces that self-
intersect, clips that miss, distances that are
"almost" right
Reproducibility: the same input must give the
same number, today and in six months
If you can explain why a model is offset,
mirrored or at the wrong scale — not just that it
looks wrong — you will do well here.
Because the output feeds real surgical planning,
correctness is not negotiable. A wrong
millimetre is a clinical problem, not a bug report.
This role is probably not for you if you want
fully predefined tickets, a fixed roadmap, or to
implement tasks without understanding the
domain. Hospitals and real clinical data change
priorities; we adjust as a team.

What We're Looking For

Python engineering (the core of this role)

• Clean, modular, strictly typed Python. We run

mypy --strict, ruff, pytest, pre-
commit — and deep, technical code reviews.

• Confident with NumPy / SciPy and array-

heavy numerical code.

• Ability to own a feature end-to-end: from

understanding the clinical problem to
validating the result on real data.

3-D medical imaging fundamentals
-Understanding of origin, spacing and
direction, and the difference between voxel
and physical coordinates.

-Reading DICOM series correctly, validating
RAS orientation, resampling without silently
corrupting geometry.

-Comfort debugging complex geometric
problems: distance, intersection, clipping,
surface generation from labels.

Toolchain
-SimpleITK, VTK or PyTorch experience is a
strong plus. If you haven't used them, you
must be genuinely comfortable diving deep
and learning independently — we mentor, but
initiative is expected.
-Docker, Git, GitHub Actions, structured
logging.

Ways of working
-Clear technical communication — you can
explain your reasoning and defend a design.
-Ownership over micromanagement. Good
written and spoken English.

What You'll Do

• DICOM & SimpleITK: read series correctly,

validate RAS origins and directions, resample
safely, maintain voxel ↔ physical transforms.

• VTK & geometry: generate surfaces from

labels; implement distance, intersection and
clipping algorithms; export reliable STL
markers and geometry artifacts.

• Measurement & QA logic: build robust

procedural tools for radiologists and handle
the edge cases real hospital data produces.

• Pipeline reliability: performance tuning,

structured logging, and making sure
pipelines survive messy multi-scanner
datasets.

Example challenges

• Build a robust distance-measurement tool

between anatomical segmentations, with
exportable nearest-point markers.

• Extract centerlines and anatomical landmarks

from noisy CT scans.

• Enforce consistent spacing / origin / direction

across multi-scanner datasets.

If these sound exciting rather than
overwhelming, you'll likely enjoy this role.

Nice to Have

• VTK surface / volume rendering experience.
• ML segmentation metrics and tooling (Dice,

Hausdorff, nnU-Net, PyTorch).

• GPU acceleration and large-volume

performance work.

• Background in radiology, biomedical

engineering or clinical imaging.

Our Stack
Python 3.13+ · SimpleITK · VTK · NumPy / SciPy

• Pydantic · PyTorch · nnU-Net · pytest · mypy

(strict) · ruff · pre-commit · Docker · GitHub
Actions · Git LFS · Sentry

What We Offer
Competitive salary, optional stock options, and
a clear path toward senior-level ownership and
technical lead responsibilities. Fully remote,
small team, no corporate theater.

How to Apply

• CV or LinkedIn
• GitHub or portfolio — we value real code
• A short note: why does working in an

evolving product environment excite you?