Senior Multimodal ML Engineer - Vision, Text & Data
Posted Updated
Description
Mollkom in Riyadh, Saudi Arabia, seeks a machine learning professional to design VLM and OCR pipelines for rich product catalogs. You will build evaluation sets across languages and categories, distinguishing visible facts from uncertain inferences while optimizing quality, cost and latency.
The role is hybrid in Riyadh, offering opportunities to turn research into observable production services and collaborate across data, engineering and product teams.
Design VLM and OCR pipelines for product attributes and structured catalog data. Build evaluation sets spanning image quality, languages and product categories. Distinguish visible facts, retrieved information and uncertain inference. Experiment with model selection, preprocessing and adaptation, measuring quality, cost and latency. Experience in computer vision or multimodal ML. Proficient in Python and PyTorch. Knowledge of data labeling, evaluation and error analysis. Ability to deploy models as production services.Skills
Sign in to read and join the discussion.