About the Role
We are looking for a hands-on Data Scientist with strong expertise in recommendation systems, classical Machine Learning, and Deep Learning.
You will work on complex ML problems involving large-scale, multimodal, and partially unlabeled data, including text, images, signals, and tabular datasets. The role requires strong modeling depth and the ability to own the full ML lifecycle — from exploratory data analysis and data validation to model development, evaluation, and production deployment.
This position is ideal for someone who understands not only how to use ML frameworks and libraries, but also the mathematical and algorithmic principles behind the models they build.
What You’ll Do
Design, develop, validate, and deploy Machine Learning and Deep Learning models end-to-end
Build and improve recommendation systems focused on user, product, and content similarity
Develop approaches for similarity matching across heterogeneous and multimodal data
Work with large volumes of structured and unstructured data, including text, documents, images, signals, and tabular datasets
Perform exploratory data analysis and validate the quality and suitability of training data
Translate business challenges into practical ML solutions and modeling strategies
Develop solutions for datasets with limited or missing labels
Explore advanced approaches such as contrastive learning, embeddings, pseudo-labeling, and representation learning
Design and evaluate similarity metrics when standard distance functions are not sufficient
Build production-quality data pipelines and model implementations
Fine-tune and adapt pretrained models for domain-specific use cases
Collaborate with engineering and business stakeholders to bring models into production
What We’re Looking For
Machine Learning & Data Science
Strong foundation in classical Machine Learning, statistics, and mathematics
Deep hands-on experience developing ML models, not only designing high-level architectures
Experience with supervised, unsupervised, and transfer learning
Strong understanding of recommendation systems and similarity-based modeling
Experience with embeddings, representation learning, and similarity search
Ability to select and justify modeling approaches based on business and data constraints
Experience with time series analysis, including at least one approach beyond standard off-the-shelf solutions
Deep Learning
Strong understanding of neural network architecture and internal mechanics
Practical experience with CNNs and Transformers
Understanding of layer design choices and their impact on model behavior
Strong knowledge of loss functions and when to use different approaches
Understanding of activation functions, gradient flow, and common edge cases such as dying ReLU
Experience with computer vision, pretrained models, and fine-tuning
Familiarity with techniques such as contrastive learning is highly desirable
Data & Algorithms
Experience validating training data quality and defining data validation methodologies
Experience working with unstructured data such as documents, PDFs, images, and text
Strong algorithmic and problem-solving skills
Solid understanding of classical algorithms, including dynamic programming, greedy approaches, and optimization problems
Ability to work with large-scale datasets of 100GB+ in cloud environments
Engineering & Cloud
Strong Python skills and ability to write clean, production-quality code
Experience building data pipelines and implementing ML solutions in production
Hands-on experience with GCP or Azure; experience with both is a plus
Familiarity with MLOps practices, deployment, and model monitoring is an advantage
Nice to Have
Experience with multimodal Machine Learning
Experience working with unlabeled or weakly labeled datasets
Knowledge of pseudo-labeling and advanced representation learning techniques
Experience with large-scale recommendation engines
Familiarity with producti
