Internship - Automated Data Quality & News-Driven Outlier Detection

Capital Fund Management (CFM); Paris; Sur place; EUR 20 000 - 31 000; Plein temps

Résumé du poste

Capital Fund Management (CFM) is seeking a Master’s student in data science to join the Equity Data Referential team in Paris. The role focuses on automating outlier review using AI and News data, improving data quality across non-production stock timeseries, and reducing manual review effort.

Applicants should be proficient with large datasets, SQL, and financial time series concepts, and be fluent in French and English.

Qualifications

• Pursuing a Master's degree in data science or a related field.
• Strong data analysis, problem-solving and automation skills.
• Comfortable working with large, high-dimensional datasets and AI tools.
• Fluent in both French and English.

Responsabilités

• For each outlier detected by the existing algorithm, develop an automated approach that leverages AI tools and historical News datasets to reduce the number of false positive issues.
• Design and implement methods to correlate detected outliers with contextual information extracted from News data, including sentiment and corporate events, to explain or validate market movements.
• Ensure the data quality of your results by benchmarking them against historical price validations that were previously performed manually.
• Address the challenges of working with large-scale, feature-rich News datasets, including sentiment scores and other signals, through efficient data processing and feature selection.
• Contribute to the improvement of the team's data quality pipeline for non-production stocks.
• Explore how AI-driven approaches can be integrated into the existing outlier detection and validation workflow.
• Help build reusable, scalable tooling to support the transition of products into production with higher confidence and lower manual effort.

Connaissances

Data science

Data analysis

Automation

Large datasets

French & English

Team spirit

News data processing

SQL

Financial time series

Corporate actions

Data quality frameworks

Formation

Master's degree in data science or related field

Outils

Python

Description du poste

Founded in 1991, we are a global quantitative and systematic asset management firm applying a scientific approach to finance to develop alternative investment strategies that create value for our clients.

We value innovation, dedication, collaboration, and the ability to make an impact. Together, we create a stimulating environment for talented and passionate experts in research, technology, and business to explore new ideas and challenge existing assumptions.

YOUR ROLE

The Equity Data Referential team plays a central and transversal role within CFM and emphasizes the values of integrity and excellence. We guarantee the reliability of the equity reference data that feeds our research and production systems, ensuring that the timeseries used across the firm are accurate, clean, and trustworthy.

We maintain a large universe of non-production stocks, each with historical timeseries of prices, number of shares, and corporate actions (CACS). Today, data quality issues, such as price jumps, are only investigated when a product is about to move into production. At that point, operators must manually review each outlier and decide whether to flag it as a valid event, such as a legitimate price jump, or fix it, such as correcting the timeseries with a valid price. This process is time-consuming and generates a high number of false positives.

Your mission will be to leverage AI tools and historical News datasets to automatically distinguish real data issues from legitimate market events, so that operators can focus only on the cases that truly require human attention.

KEY RESPONSIBILITIES

• For each outlier detected by the existing algorithm, develop an automated approach that leverages available AI tools and historical News datasets to reduce the number of false positive issues, allowing operators to concentrate only on genuine data quality problems.
• Design and implement methods to correlate detected outliers, such as price jumps, with contextual information extracted from News data, including sentiment and corporate events, to explain or validate market movements.
• Ensure the data quality of your results by benchmarking them against historical price validations that were previously performed manually.
• Address the challenges of working with large-scale, feature-rich News datasets, including sentiment scores and other signals, through efficient data processing and feature selection.
• Contribute to the improvement of the team's data quality pipeline for non-production stocks.
• Explore how AI-driven approaches can be integrated into the existing outlier detection and validation workflow.
• Help build reusable, scalable tooling to support the transition of products into production with higher confidence and lower manual effort.

YOUR SKILLS

• You are pursuing data science studies, at Master's level or equivalent, with a focus on dataset quality.
• You have an aptitude for data analysis, problem investigation, and the automation of manual processes.
• You are comfortable working with large, high-dimensional datasets and leveraging existing AI tools to extract value from them.
• You are comfortable in both French and English.
• We are looking for motivated, curious, rigorous personalities with a strong team spirit.
• Experience with processing large text/News datasets.
• Good knowledge of SQL, and familiarity with financial timeseries and corporate actions.
• Experience with data quality frameworks and validation methodologies.

EQUAL OPPORTUNITIES STATEMENT

We are continuously striving to be an equal opportunity employer and we prohibit any discrimination based on sex, disability, origin, sexual orientation, gender identity, age, race, or religion. We believe that our diversity, breadth of experience, and multiple points of view are among the leading factors in our success. CFM is a signatory of the Women Empowerment Principles.

FOLLOW US

Follow us on Twitter or LinkedIn or visit our website to find out more about CFM.