ML Engineer with Distillation experience

Apply
Apply

Share

successfully icon

Successfully

The vacancy has been successfully added to favorites

location icon

Bucharest, Romania

specialization icon

AI/ML

lob icon

Automotive Industry

date icon

26/02/2026

Req. VR-121211

Apply
Project description

High-tech project at the intersection of GPU-accelerated ML inference and real-time graphics. The goal is to build fast, reliable, cost-efficient inference pipelines for visual computing and integrate ML components into real-time rendering / engine workflows. A key focus is model distillation and compression to reduce latency, memory footprint, and infrastructure cost, with production deployment using standard engineering practices.

Responsibilities
bullet icon

Build and optimize GPU inference pipelines for low latency and high throughput

bullet icon

Implement model distillation / compression to make models faster and cheaper

bullet icon

Profile and tune performance across CPU and GPU (latency, throughput, memory)

bullet icon

Integrate ML into real-time graphics workflows / pipelines (engine-side integration when needed)

bullet icon

Maintain production readiness: reproducible builds, basic CI/CD, containerized deployment

Skills

Must have

bullet icon

Senior-level ML/Inference Engineer (3-5+ years), able to work independently

bullet icon

Strong deep learning + CNN background, practical experience shipping models to inference

bullet icon

Distillation / compression experience (any solid KD / compression practice)

bullet icon

Strong Python + PyTorch OR equivalent (enough to implement training/inference and debug)

bullet icon

Strong GPU inference/performance mindset: CUDA fundamentals, profiling, optimization approach

bullet icon

Solid software engineering skills (clean code, testing basics, Git, collaboration)

Nice to have

bullet icon

Hands-on with TensorRT / ONNX Runtime / Triton (any of them)

bullet icon

Quantization / mixed precision / operator fusion experience (any subset)

bullet icon

Experience integrating ML into graphics or engines (Unreal/Unity, rendering pipeline basics)

bullet icon

Kubernetes/Docker in production, observability/telemetry practices

bullet icon

Familiarity with image quality metrics (SSIM/PSNR/LPIPS)

Other
seniority icon

Languages

English: B2 Upper Intermediate

seniority icon

Seniority

Senior

Bucharest, Romania

Req. VR-121211

AI/ML

Automotive Industry

26/02/2026

Req. VR-121211

Apply for ML Engineer with Distillation experience in Bucharest

*Indicates a required field

Under the terms of your specific consent or to perform our obligations under a contract with you, as applicable, we, Luxoft Holding Inc. will manually and electronically process your personal data, specifically your first name, last name, phone number, e-mail address and other data you provide us through this form.


Within this context, we process personal data only for the specific purpose(s) indicated in the individual consent language or other notices provided below.


We will – insofar as reasonably necessary for the purpose you have agreed to and within the scope of applicable laws – transfer your personal data to other entities within the Luxoft Group and to the group of third party recipients listed in our Privacy Notice. Such Recipients can be located outside the European Union (EU) and/or the European Economic Area (EEA) (“Third Countries”). The Third Countries concerned, e.g. the USA, may not have the level of data protection that you enjoy e.g. under the GDPR. This can result in disadvantages such as an impeded enforcement of data subjects’ rights, a lack of control over further processing and access by state authorities. You may only have limited legal remedies against this. Insofar our transfer of your personal data to recipients in Third Countries is not covered by an adequacy decision of the EU Commission, we achieve an adequate level of data protection as further detailed out in our Privacy Notice.


With your consent, we personalise marketing communications to you by way of carrying out marketing research analysis, analysing the surfing-behaviour of our website visitors and to adjust it to their detected tendencies, as well as to plan more efficient future marketing activities. This personalised marketing does not include any automated decision-making activities.


Further information on how we process personal data in general is available in our Privacy Notice. You may withdraw any given consent at any time. The withdrawal of your consent(s) will not affect the lawfulness of processing before its withdrawal. For any request in this context, please e-mail us at: DPO@luxoft.com.


Before uploading CV or any other information to this website, to learn more about your obligations and restrictions arising from the use of this website, please read our Terms of Use.