Job Description

Portcast is a venture-backed startup which predicts global trade flows to help logistics and shipping companies become more profitable. We are a predictive analytics company that offers a fast-paced, innovative environment where you will be empowered to sell our AI-product to C-level executives. We are customer-obsessed and are constantly working to provide our customers access to actionable and insightful data to build resilient supply chains.

Our mission is to transform international supply chains to be more resilient by helping logistics companies realise the full potential of their data. We cater to both shipping lines and cargo airlines. This covers 90% of the world trade volume that travels via ocean and 35% of world trade value that travels via air. We use proprietary machine learning algorithms and real-time external market data (such as economic indices, marine weather, satellite-based data, etc) to predict how much cargo will be shipped, when it will arrive and deliver actionable insights.

About the role

Ocean transportation data is segregated across multiple sources like ocean carriers, satellite, ports, etc. In order to reach exceptional data quality, Portcast has set up processes to deep dive into completeness, correctness and accuracy of the data at each step of the ocean movement. The Data Science trainee will work with cross-functional teams (data science, software development and business) to identify areas where model features can be improved to deliver better accuracy and granularity in the event of contingencies.

\n

What You'll Do:

• R&D: Machine Learning - You'll work closely on identifying of the new derived features based on Speed on Ground, Trajectory prediction, Vessels metadata etc. Think of all the potential features that might affect global container ship movement (e.g. the latest Suez Canal blockage, congestion in Chinese ports due to the Delta outbreak, typhoon In-Fa in East China Sea).
• Prototypes of new models with different cross validation accuracy scores (based on time, geography etc) compared against the benchmark.
• Anomaly detection based on vessel behaviour and other port, route based features.
• Exploratory Data Analysis: Exploration of AIS (Automatic Identification Systems) data with millions of geolocation records along with vessels metadata.
• Review, Update of Ranking algorithm for carrier schedules. Delay is always relative to the universe of the schedule that the carrier is operating. Hence, we use dynamic carrier selection feature that is powered by the ranking algorithm.
• Identification of the features from each dataset and understanding how those features will support existing Portcast models.
• Impact Analysis: Storytelling, Data visualization, Interpretability.
• Identification of different flags/alerts based on predictions (cyclone, vessel route change, anomalies).
• Compelling data stories on how predictions can be helpful, able to answer questions like below:

1) What is an average delay expected at port CNYTN because of port congestion that was caused by the Typhoon In-Fa?

2) What would be the new ETA if vessel plan to avoid Suez canal / take a detour around Cape of Good Hope?

3) What would be the new ETA if vessel skips a port?

4) What is the impact on CO2 emissions if the vessel were to take a detour / sail faster?

Requirements:

• Currently pursuing a Bachelor’s or Master’s degree in Computing, Statistics, Mathematics, Machine Learning or AI, Computer Science, Engineering, or a related field.
• Experience with machine learning models with deep statistical knowledge.
• Strong analytical skills with experience handling large datasets and data visualization.
• Proficiency in Excel, SQL, and basic Python (Pandas, NumPy) is preferred.
• Knowledge AI and machine learning applications within logistics and supply chain optimization.
• Strong communication skills and ability to translate data-driven insights into meaningful recommendations.

What's In It For You:

• Hands-on experience with AI-driven supply chain models and real-world machine learning applications.
• Exposure to both predictive modeling and data quality assurance in a fast-growing technology-driven company.
• Opportunity to work on time-series forecasting, anomaly detection, and AI OCR.
• Mentorship from experienced data scientists and industry experts.

\n

Please mention the word **AFFLUENCE** and tag RNDQuMjI2LjEyMi4z when applying to show you read the job post completely (#RNDQuMjI2LjEyMi4z). This is a beta feature to avoid spam applicants. Companies can search these words to find applicants that read this and see they're human.

Job Tags

Remote job, Full time, Traineeship,

Similar Jobs