About Puma Energy

At Puma Energy, data is at the core of how we transform our global operations and deliver value across the energy value chain. Our Data & Analytics team is building a modern cloud-native data platform on AWS and Databricks, enabling trusted, scalable data products that power analytics, AI, and business-critical decision making.

We are looking for a Senior Data Scientist who is fundamentally a builder — someone who personally designs, ships, and owns machine learning and agentic automation systems in production, and who also raises the technical bar for the wider team through hands-on mentorship and sound methodology. This is not a purely strategic or advisory role: you will be in the code, in the data, and in the model — while also guiding other data scientists and engineers on the right approach to take. You will partner closely with our Data Engineering function (which builds our scalable data products on Databricks/AWS/Spark) and with commercial, retail, supply, and trading stakeholders.

This role goes beyond model-building in a notebook. You will:

• Own data science products end-to-end — from framing an ambiguous business problem to shipping and monitoring a production model.
• Build and deploy applied agentic automation solutions that augment or replace manual analytical/operational workflows.
• Guide teams on the right technical approach — modeling technique, architecture, and productionization strategy — without taking on formal people management.
• Partner with business stakeholders to translate complex problems into scalable, measurable data science and automation solutions.
• Continuously raise the bar on experimentation rigor, reproducibility, and production discipline across the team.

Responsibilities

Build & Ship Production Data Science

• Design, build, and maintain machine learning models and data science solutions using Databricks, Spark (PySpark), Python, SQL, and AWS.
• Own the full ML lifecycle: problem framing, data preparation, feature engineering, model training and evaluation, deployment, monitoring, and retraining.
• Build reusable modeling frameworks, curated feature sets, and self-service data science assets for forecasting, optimization, segmentation, and predictive use cases.
• Write clean, maintainable, production-grade code with strong testing, version control, and code-review discipline.

Applied Agentic Automation

• Design and build applied agentic automation workflows (e.g., LLM/agent-based systems that plan, retrieve, reason, and take action) to automate analytical, operational, or reporting processes.
• Evaluate and select appropriate agentic frameworks and tooling, balancing reliability, cost, and maintainability for production use.
• Establish guardrails, evaluation, and monitoring practices for agentic systems deployed into business workflows.

Solution Design & Technical Guidance

• Partner with business stakeholders to understand requirements and translate them into scalable data science and automation solutions.
• Guide and mentor other data scientists and engineers through design reviews, pairing, and code/methodology review — setting the technical direction without formal management authority.
• Evaluate architectural and modeling trade-offs and influence the team's technical standards and best practices.
• Drive solution design from concept through production deployment.

Engineering Excellence

• Build production-grade solutions with a strong focus on quality, testing, observability, and reliability.
• Implement automated model validation, monitoring, and retraining pipelines in partnership with Data Engineering and MLOps.
• Troubleshoot production issues and continuously improve model performance and developer experience.

Business Partnership

• Collaborate with business stakeholders, product owners, analysts, architects, and engineers to solve high-impact business problems.
• Communicate technical concepts and results clearly to technical and non-technical audiences.
• Take end-to-end ownership of solutions from discovery and design to production support.

Required Skills & Qualifications

Core Technical Skills

• 6–10 years of hands-on experience in Data Science / Machine Learning, with a proven track record of building, deploying, and maintaining ML solutions in production — not solely prototyping or advisory work.
• Strong, demonstrable hands-on experience with Databricks is essential for this role — this is a non-negotiable requirement, not a nice-to-have.
• At least 1 year of hands-on experience building applied agentic automation solutions (e.g., LLM agents, agentic workflows, tool-using AI systems) that have been deployed or piloted in a real business context.
• Strong expertise in Python, SQL, and at least one ML framework (TensorFlow, PyTorch, or scikit-learn).
• Hands-on experience with Apache Spark (PySpark) and cloud-native data platforms on AWS.
• Strong understanding of the end-to-end ML lifecycle: experimentation, validation, deployment, and monitoring at scale.
• Strong software engineering fundamentals including Git, CI/CD, testing, and code quality.
• Excellent communication skills with the ability to influence stakeholders, mentor peers, and drive technical discussions — without requiring formal people-management experience.

Nice to Have

• Databricks or AWS Certifications.
• Experience with MLOps tooling (MLflow, Airflow/Prefect) and Infrastructure as Code (Terraform).
• Experience with experimentation / causal inference and A/B testing methodology.
• Domain exposure to energy, commodities, downstream fuel/retail, or trading analytics.
• Data visualization skills (Power BI, Tableau, Plotly/Dash).

What Success Looks Like

• Own and deliver production-grade data science and agentic automation solutions powering business-critical decisions.
• Enable at least one applied agentic automation workflow that measurably reduces manual effort or improves decision speed.
• Visibly raise the technical bar of the team through mentorship, design reviews, and best practices — while remaining hands-on.
• Build strong partnerships with business stakeholders and deliver measurable business outcomes.
• Continuously improve model reliability, scalability, and cost efficiency on the Databricks/AWS platform.

Why Join Puma Energy

• Build on a modern AWS + Databricks Lakehouse platform used across a global enterprise.
• Work on challenging, high-impact data science and agentic automation problems with real business impact.
• Own solutions end-to-end with autonomy and accountability — a hands-on builder role, not a purely strategic one.
• Grow your career through technical leadership and mentorship, on an individual-contributor track, without being forced into people management.