The role

Remote · Any time zone · English C1+

You will own production ML systems end to end for a large-scale e-commerce platform: data processing, model training, deployment, and continuous improvement of accuracy, reliability and cost. You’ll work in a small, senior team alongside data scientists and an MLOps engineer.

Stack: Python, PySpark, Databricks, PyTorch/TensorFlow, AWS (SageMaker, ECS/Fargate, Lambda, S3, Glue, Athena, DynamoDB, CloudWatch).

Current priorities: modernizing large-scale data pipelines, consolidating orchestration, reducing cloud processing costs, and improving model quality through A/B testing.

What you’ll do

- Build and productionize models for recommendation, ranking, prediction and forecasting.
- Develop batch and real-time PySpark/Databricks pipelines for ingestion, features, training data and inference.
- Ship models to production with serving, APIs, containers, CI/CD, monitoring and retraining.
- Evaluate models with offline metrics and online A/B tests tied to business outcomes.
- Optimize performance, scalability and cost of ML and data workloads.
- Contribute to architecture decisions and engineering standards.

Must-haves
- 5+ years of hands-on ML engineering in industry. Research, academic, data engineering or AI architecture roles don’t count.
- Recommender systems or consumer-domain experience (at least one).
- Recommender work: collaborative filtering, implicit feedback, ALS/matrix factorization, candidate generation and ranking, cold start,



offline and online evaluation.
- Consumer domain: ML engineering in retail, e-commerce or consumer companies.
- Production deep learning. Models you designed, trained, deployed and operated with PyTorch or TensorFlow, including evaluation, monitoring and retraining.
- Large-scale distributed processing. Spark/PySpark or equivalent on real big-data workloads, including performance tuning. You should be able to state the data volumes and explain why distributed processing was needed.
- Cloud ML deployment. AWS preferred.
- Strong Python and SQL with production-quality engineering practices.
- Clear technical communication in English (C1 minimum).

Nice-to-haves
- Both recommender systems and e-commerce/marketplace personalization experience
- Databricks (Jobs, Workflows, Delta Lake)
- Airflow or similar orchestration
- MLflow or similar experiment tracking and model registry
- Docker and FastAPI-based model serving
- A/B testing against business metrics
- Latency and cloud cost optimization
- LLM, RAG or agent experience, as a complement to core ML, not a replacement for it

Not a fit if your background is mainly in
- ML limited to financial services or insurance
- Research or academia without 5+ years in industry
- Data engineering, research data science or AI architecture without hands-on model ownership
- GenAI/LLM work without production deep learning on large data volumes

Lugar de trabajo: Empleo remoto

📌 Senior Machine Learning Engineer (Argentina)
🏢 Avalith
📍 Argentina