Data Analyst

Project overview

Missions

Establish a testing protocol to measure the impact of matching rules developed by the team

Pilot testing campaigns and produce a quantitative analysis of discrepancies to adjust matching criteria

Conduct campaigns comparing two versions of the identification engine, with quantitative and qualitative analyses of the differences

Actively participate in the automation of data analysis to accelerate time to deliver

Analyze anomalies and user feedback with business experts, and build rule evolutions with the PO

Participate in the validation and non-regression of developments, in conjunction with the PO and developers

Get involved from the scoping phase (shift-left): acceptance criteria for User Stories and testing strategy

Enrich analyses with internal, external data, or via algorithms

Stack / Environment

Python for analysis and automation (pandas, numpy, notebooks), good command

Elasticsearch, SQL, good command

AWS (OpenSearch, RDS, SQS, SNS, DynamoDB, ECS), partial command

Spark, beginner level sufficient

REST Client (Postman), beginner level sufficient

Analysis protocol (exploration, representative sampling, pattern detection), good command

Agile / Scrum, good command; SAFe, beginner level

Profile

Experienced level, Master's degree (Bac+5).

Significant experience in data analysis.

Comfortable with testing methodologies on an application with uncertain results.

Ability to quickly acquire complex functional knowledge.

Good command of writing test cases and establishing a testing protocol.

Knowledge of textual analysis techniques (normalization, string distances, TF-IDF, BM25, fuzzy matching).

NLP skills appreciated, beginner level sufficient.

Strong analytical mindset, good written and oral communication, dynamic, autonomous, proactive.

3 key skills

Python Data Analysis

Tests Matching Quality

SQL Elasticsearch AWS

Work Environment

Our client is a key player in its sector. It processes very large volumes of data transmitted by external partners each year, and this data is used for high-stakes financial allocation calculations.

At the heart of the system is an internally developed matching engine that reconciles millions of incoming data lines with an internal repository. It relies on standardized identifiers and textual matching (labels, names, associated entities). The product is continuously evolving with two objectives: to identify more and more data and to reduce the margin of error.

You will join an agile product team composed of a lead dev, two developers, and a data analyst.

Project details

• Start: 28/10/2026
• Duration: 12 mois renouvelable
• Remote: hybrid
• Location: Neuilly-sur-Seine, Île-de-France, FR
• Languages: French