Lead Data Scientist

About Vantora:

Vantora builds enterprise-grade AI companies alongside Fortune 500 partners to solve their toughest problems. We bring together world-class entrepreneurs, engineers, and operators to turn complex industry challenges into new companies, combining startup speed with the data, infrastructure, and market access of established enterprises. With over $100M in backing from Silversmith Capital Partners, we're building the next generation of AI-native businesses across the physical world.

About the Opportunity

We’re building a next-generation AI model that will transform how the exploration energy industry operates. This venture, developed in partnership with one of the world’s leading oil and gas companies, will leverage LLMs, machine learning, and domain-specific data to help scientists make smarter, faster, and more profitable decisions.

As Lead Data Scientist, you'll be at the core of this effort, designing, training, and deploying models that turn complex, messy, real-world oil and gas data into actionable intelligence. You'll work closely with product, engineering, and domain experts to prototype, validate, and productionize AI solutions that directly impact how our users plan, produce, and operate.

In This Role, You Will

• Design and build ML models for prediction, optimization and pattern detection across very large, multi-source datasets (time-series, sensor, operational and transactional data).
• Work with product managers, engineers, data engineers and domain experts to define use cases, gather requirements and turn business problems into quantitative ones.
• Build and maintain data pipelines that support model training, evaluation and deployment at scale.
• Prototype quickly, testing different approaches (regression, time-series forecasting, clustering, reinforcement learning, deep learning) to find what works best.
• Improve model performance through rigorous experimentation, feature engineering and error analysis.
• Partner with the AI/LLM engineering team to build predictive and generative components into end-user workflows.
• Explain findings clearly to technical and non-technical audiences so model outputs lead to business decisions.
• Set best practices for reproducibility, versioning and documentation as the product scales.

What You Bring

• 6+ years of experience in applied data science, machine learning or advanced analytics.
• Deep hands-on experience with very large datasets (billions of records or multi-terabyte scale), including distributed processing, performance tuning and working with data that’s messy, incomplete or comes from many sources.
• A track record of building and deploying models in production, ideally in B2B SaaS or an enterprise environment.
• Expertise in Python, SQL and common ML frameworks (scikit-learn, PyTorch, TensorFlow, XGBoost, etc.).
• Strong data wrangling, feature engineering and exploratory analysis skills across structured and unstructured data.
• Experience with cloud ML pipelines (GCP, AWS or Azure) and data infrastructure tools (Databricks, Spark, Airflow, etc.).
• The ability to design experiments, interpret the results and use data to guide decisions.
• Strong communication skills, including putting technical work in context for executives, operations teams and business stakeholders.

Nice to Have

• Preferred: experience in oil and gas, energy or another heavy industrial setting, especially with operational data like production, drilling, equipment or sensor/SCADA data.
• Time-series forecasting, predictive maintenance or anomaly detection work.
• Exposure to physics-informed ML or hybrid models that combine engineering principles with data-driven methods.
• Experience in an early-stage, zero-to-one product environment.

Preferred Locations: Houston, TX; open to Los Angeles, CA or Remote - 50% travel required