Careers

Machine Learning Engineer

Closed

Iași / Remote (Romania) · Full-time

You own the model side of our AI work: the training, evaluation, and inference pipelines behind features like Censory’s PII detection. Less notebook, more production: models that run reliably, cheaply, and fast enough to sit in front of real users.

This role is currently closed

We're not hiring for this role right now. We're just starting to build the team, and we'll reopen roles as we grow. Browse all our roles, or reach out anyway if you think you'd be a great fit for the team.

What this role involves

Build and maintain training, fine-tuning, and evaluation pipelines that other engineers can rely on
Take models from experiment to production inference: packaged, monitored, and cost-aware
Optimize latency and cost across cloud GPUs and local / quantized model serving
Partner with the AI and platform engineers to turn model outputs into shippable product features

What we look for

3+ years in ML / applied ML with models you have taken all the way to production
Strong Python and a modern ML stack (PyTorch, Hugging Face, or equivalent)
Solid grasp of evaluation, data quality, and the failure modes of ML in the real world
Comfortable owning inference performance, not just training metrics

Nice to have

Quantization and local-model serving (GGUF, llama.cpp, ONNX, TensorRT)
Computer vision or document/OCR pipelines
MLOps tooling and reproducible training infrastructure