Senior Deep Learning Algorithm Engineer
We are looking for a Senior ML Algorithm Engineer to lead the development and optimization of machine learning models for challenging real-world problems.
In this role, you will work hands-on across the full model-development lifecycle: understanding the task, designing and adapting algorithms, building effective training strategies, analyzing data and failure modes, and improving model quality and efficiency through rigorous experimentation.
This is an algorithm-focused role for someone who enjoys getting deeply involved in the details of model training and optimization. You will not simply operate an existing training infrastructure—you will investigate open-ended problems and develop practical solutions involving model architecture, data and sample selection, training objectives, optimization methods, and evaluation
What will your job look like:
- Work on the semantics of road objects, using harvested tabular data as the primary input to our algorithms — turning large-scale, real-world observations into models that capture object meaning, attributes, and behavior on the road network
- Work directly with training code to design and optimize ML and deep learning algorithms, adapting architectures, objectives, loss functions, sampling strategies, and optimization methods
- Own the end-to-end training process, from data preparation and dataset composition through training, validation, and evaluation.
-
Analyze model behavior and failure modes, lead controlled experiments, and translate insights into measurable improvements in model quality, efficiency, and scalability.
All you need is:
- 3+ years of experience in algorithm engineering using deep learning.
- Strong hands-on experience designing, training, evaluating, and optimizing deep learning models.
- Strong programming skills in Python, with hands-on experience developing and training models using PyTorch
- Experience processing large-scale datasets with Pandas, Spark, or equivalent technologies, in a cloud environment such as AWS or GCP.
-
Strong analytical and problem-solving skills, with a practical, hands-on, and ownership-driven mindset.
- Experience with Polars and DuckDB- an advantage
