ML/AI Engineer (Databricks)

Elios, Inc.; Brasil; Remote; BRL 180,000 - 300,000; Full time

Job summary

Elios, Inc. is seeking an experienced ML engineer to build and deploy ML models on a modern Databricks data platform.

You will transform messy enterprise sources into features and AI-powered applications that run in production, touching pipelines, models, and services used by real users. You'll work end-to-end across data engineering and ML pipelines, standing up MLflow workflows, and applying LLM or retrieval techniques when they fit the problem.

Qualifications

• 3 to 5+ years building ML or AI systems in production.
• Strong Python fundamentals.
• Hands-on experience with Databricks (Spark and Delta Lake).
• Comfort across the ML lifecycle: feature engineering, training, deployment, monitoring.
• Working knowledge of MLflow or similar experiment tracking tools.
• Strong SQL skills and data pipeline understanding.
• Exposure to LLM-powered or retrieval-based applications.

Responsibilities

• Build, train, and validate ML models on Databricks.
• Engineer features and data transformations for enterprise data.
• Stand up MLflow-based workflows for experiments and model registry.
• Deploy models as batch jobs or real-time endpoints with monitoring.
• Collaborate with data engineers to ensure reliable data feeding models.
• Incorporate LLM or retrieval techniques when appropriate.
• Maintain strong engineering practices including version control, testing, and CI/CD.

Skills

Python

Databricks

Spark

SQL

MLflow

LLM/Retrieval

CI/CD

Version control

Tools

Databricks

Spark

Delta Lake

MLflow

Job description

Job Description

About the Role

You'll build and ship machine learning and AI systems on top of a modern Databricks data platform. The work sits close to the data, turning messy enterprise sources into features, models, and AI-powered applications that run reliably in production, not just in a notebook.

Our client helps energy and heavy industry enterprises modernize how they work with data, moving legacy systems onto Databricks and the lakehouse, then building intelligence on top. This is hands-on, full-lifecycle work. You'll touch pipelines, models, and the services that put predictions in front of real users.

What You'll Do

Build, train, and validate ML models on Databricks, with real attention to evaluation, generalization, and where the model will actually be used

Engineer features and data transformations in Spark and SQL from enterprise sources that are rarely tidy

Stand up MLflow-based workflows for experiment tracking, model registry, and reproducible training runs

Deploy models as batch jobs or real-time endpoints, then monitor drift, latency, and cost

Work closely with data engineers on the lakehouse layer so the data feeding your models is reliable and well modeled

Bring in LLM and retrieval techniques where they fit the problem, and skip them where they don't

Strong engineering habits matter here, including version control, testing, code review, and CI/CD

Qualifications

3 to 5+ years building ML or AI systems in production, with strong Python fundamentals

Hands-on experience with Databricks, including Spark and Delta Lake (Azure Databricks is a plus)

Comfort across the ML lifecycle: feature engineering, training, validation, deployment, and monitoring

Working knowledge of MLflow or a comparable experiment tracking and model registry tool

Strong SQL skills and a real understanding of how data pipelines feed models

Exposure to LLM-powered or retrieval-based applications, or genuine curiosity to grow into them

Professional working English, since collaboration happens across a distributed team