Senior Data Scientist - Maritime Risk and Compliance

At Kpler, we simplify global trade information and provide valuable insights. Founded in 2014, our goal is to help over 10,000 organisations by offering the best intelligence on commodities, energy, and maritime through a single platform. Working at Kpler means you'll be a key player in turning complex data into strategic resources for our clients. Your role involves creating data-driven stories that empower clients in their industries. Your expertise helps Kpler navigate markets successfully. Your journey starts here, where innovation meets impact. Join our team of 500+ talented people from 35+ countries worldwide.

Purpose of the role: Maritime domain awareness and commodity tracking involve the effective fusion of fragmented information that compose the complex commodity flows landscape. When combined efficiently, vessel tracking data and trading history, commodity supply and demand, proprietary trading data, and machine learning algorithms provide an understanding of events at ports and on open seas at different levels of granularity. Rapidly evolving global geopolitical developments and tensions significantly affect trading, disturbing the supply chain and altering commodity flow patterns which brings more uncertainty about the modelling of these phenomena.

As a senior data scientist with Kpler’s Flows team, you will develop algorithms that extend our comprehension of commodities flows in space and time by devising forecasting models, predicting future trades, detecting anomalies in data and anomalous behaviour of various players.

Some of the exciting projects you will work on include:

• Driving the design, rapid experimentation, development, testing and deployment of data science models for flow forecast models and anomaly detection;
• Optimising and fine-tuning models in production, overseeing the continuous monitoring of deployed models, and effectively handling model and data drift promptly;
• Building robust pipelines for integrating data from diverse sources, including big geospatial data, ship mobility data, and document recognition;
• Researching and identifying methods, data sources, and features that will drive business impact and improve models’ accuracy in the current scope of ever-changing world-scale commodity trading.

As a senior data scientist, you will:

• Devise efficient solutions to tackle ML/Big Data challenges using relevant, up-to-date methods and technologies.
• Work across the stack to deliver new features end-to-end, from prototyping, to deployment and caring for data drift in production.
• Ensure optimal, cost-effective design decisions that improve performance and overcome scalability limits.
• Own meaningful parts of our service, demonstrating the ability to lead projects independently, have an impact, and grow with the company.
• Identify opportunities for novel projects and liaise with product teams to advance ideas into value-adding features.
• Actively share knowledge and document insights, effectively communicate complex concepts and analysis to technical and non-technical audiences, aiming to support continuous team improvement and drive collaboration.
• Act as a mentor for our junior data scientists, helping to accelerate their growth; you will act as the Tech Lead on some projects.
• Be part of a vibrant Machine Learning community in Kpler, tackling the whole spectrum of ML problems.

Our Machine Learning tech stack includes: Python, ML libraries (TensorFlow, pytorch, scikit-learn, transformers, XGBoost, ResNet), geospatial libraries (shapely, geopandas, rasterio), CV libraries (scikit-image, OpenCV, yolo, Detectron2). AWS, Postgres, Apache Airflow, Apache kafka, Apache Spark.

Mandatory requirements:

• You have at least 5 years of experience in the DS role, deploying models into production;
• You have proven experience delivering end-to-end ML solutions that produce business value.
• You are proficient in Python.
• You have expert knowledge of at least one cloud computing platform (preferably AWS).
• You are fluent in English.

Nice to haves but not mandatory:

• You have expertise on applications focusing on geospatial data and mobility analytics (highly desirable).
• You have proven experience with big data technologies, specifically Spark and Kafka.
• You have experience working with state-of-art ML pipeline technologies (such as MLflow, Sagemaker...) or building a ML pipeline by yourself (Docker, Kubernetes, Paperspace, Airflow...).
• You have a Ph.D. in a quantitative field (computer science, mathematics, physics, engineering...).
• You are familiar with the shipping industry and commodity trading.
• You are comfortable with software engineering best practices.
• You value code simplicity, performance and attention to detail.
• You have experience working in an international environment.

We're a dynamic company dedicated to nurturing connections and innovating solutions that tackle market challenges head-on. If you're driven by customer satisfaction and thrive on turning ideas into reality, then you've found your ideal destination. Are you prepared to embark on this exciting journey with us?