Machine Learning Engineer

KAVAK is Latin America’s fastest-growing e-commerce platform specializing in the purchase and sale of pre-owned vehicles. We pride ourselves on working smart and making things happen, delivering the easiest, safest, and most transparent experience for buying or selling cars. Our ambition is global, and we are expanding our footprint into the vibrant GCC region as part of our mission to transform the automotive market worldwide. With over 180 team members in the GCC and growing, we are committed to attracting top talent to build the technology, processes, and products that position us as the leading player in the pre-owned car market.

Key Responsibilities
- Build and maintain the listing-to-SKU mapping engine using rules, text matching, classifiers, confidence thresholds, and human review queues.
- Develop an image-based car identification model to classify make, model, year, generation, and trim for listings lacking usable text.
- Manage and optimize the local pricing algorithm by monitoring performance, running experiments, improving accuracy, reducing bias, and reporting outcomes.
- Design backtests, holdout sets, and audit samples to measure model accuracy and provide segmented reports.
- Productionize models through scheduled pipelines with monitoring, retraining, and data quality checks.
- Build and maintain data pipelines integrating data from warehouses, scrapers, catalogs, and back-office systems.
- Create supporting models for pricing strategy, including depreciation curves, price elasticity, days-to-sell, and demand forecasting.
- Convert reviewer feedback and manual overrides into training data and rules with expiration dates.
- Document models, assumptions, and results clearly for non-technical stakeholders.
- Collaborate with Pricing, Data, Product, and global teams on data access, deployment, and prioritization.

Required Qualifications
- 3 to 5 years of experience in machine learning engineering or applied data science, with at least two models deployed to production.
- Background in e-commerce, marketplaces, automotive, fintech, or pricing/valuation domains.
- Proficiency in tabular models (pricing, regression, classification) as well as text and image data processing.
- Bachelor’s degree in Computer Science, Data Science, Statistics, Mathematics, Engineering, or a related field.
- Master’s degree in Machine Learning or Data Science is preferred.

Technical Skills
- Expertise in Python ML stack including pandas, scikit-learn, XGBoost/LightGBM, and PyTorch or TensorFlow.
- Strong SQL skills on cloud data warehouses such as BigQuery or Databricks.
- Experience with pricing models including regression, quantile regression, and prediction intervals.
- Knowledge of text matching and entity resolution techniques like normalization, fuzzy matching, tokenization, and embeddings.
- Solid understanding of statistical evaluation methods including holdout design, backtesting, calibration, confidence intervals, and A/B testing.
- Familiarity with MLOps tools such as Docker, Airflow (or similar), experiment tracking (MLflow or Weights & Biases), monitoring, and retraining.
- Data quality assurance using automated validation tools like Great Expectations.
- Engineering best practices including GitLab CI/CD, code reviews, and testing.
- Experience with cloud platforms such as GCP or AWS.
- Computer vision skills in fine-grained image classification, transfer learning, image embeddings, and vision-language models.
- Working knowledge of LLMs and vector search technologies for structured data extraction and embedding matching.
- Pricing analytics expertise, including price elasticity and survival analysis for days-to-sell.
- Human-in-the-loop systems using labeling tools (Label Studio, CVAT) and active learning to reduce manual review.
- Model serving using REST APIs with FastAPI.
- Web scraping experience with Scrapy or Playwright, including handling site layout changes and compliance with terms of use.
- Reporting skills with dashboards built in Looker Studio, Streamlit, or Databricks.

Preferred Qualifications and Benefits
- Certifications such as Google Cloud Professional Machine Learning Engineer or AWS Certified Machine Learning Engineer.
- Learning Engineer – Associate certification and DeepLearning.AI specializations are a plus.
- Proficiency in Arabic language is advantageous.

If you are passionate about leveraging machine learning to revolutionize the automotive market and want to be part of an innovative and fast-growing team, we encourage you to apply and join us on this exciting journey.