AI & Data Science Student

Context & Challenge

Clinical trials are governed by multiple frameworks such as ICH-GCP, EU MDR, GDPR, national regulations, and more than 300 internal SOP documents. Ensuring compliance during daily operational decisions is critical, but manually searching across these sources is time-consuming and error-prone.

Project Goals

• Design an AI assistant architecture based on a closed, controlled knowledge base
• Implement a robust Retrieval-Augmented Generation (RAG) approach for document-grounded Q&A
• Ensure answers are traceable, explainable, and auditable (citations + source tracking)
• Prevent hallucinations by enforcing strict document grounding
• Provide a simple workflow to allow CERC to add/update/remove documents autonomously
• Address governance, traceability, and compliance requirements

Planned Deliverables

• A functional proof of concept (PoC) of the assistant
• Technical documentation (architecture, indexing strategy, key design choices)
• Tested real-world use cases with CERC teams
• A proposed method for continuous updates of the reference corpus

Project Objective

The project aims to:

• Monitor weather conditions (temperature, rain, wind, UV, etc.) over a configurable time window (J-x → J+y)
• Determine whether conditions are globally acceptable using tolerance logic
• Automatically send summary emails when conditions are met, while avoiding notification spam

System Architecture

The solution is built around a Python Flask backend, structured into clear functional modules:

• A Flask web application exposing HTTP routes for form submission and Excel import
• A core business logic layer that validates inputs and orchestrates the analysis
• A weather analysis module interacting with the Open-Meteo API
• An email module handling SMTP notifications
• A JSON-based cache used for persistence, state tracking, and anti-spam logic

This architecture allows the system to run both:

• on demand (via the web interface), and
• automatically (via a scheduler that reuses the last configuration).

Backend & Business Logic

A key part of the project was translating business-friendly inputs into reliable backend logic:

• Parsing and validating user inputs coming from a web form or a structured Excel file
• Converting textual business rules (thresholds, rain levels, time windows) into numerical conditions
• Building pandas DataFrames to analyze weather data day by day
• Applying a tolerance rule (up to one-third of days in a window may be non-compliant) to reduce false negatives

Weather Analysis

The backend interacts with the Open-Meteo API to:

• Geocode store locations using postal codes
• Retrieve forecast and historical weather data
• Build detailed daily DataFrames containing weather values and compliance indicators
• Decide whether a store is globally compliant across the configured time window

Automation & Anti-Spam Mechanism

To ensure reliability and usability:

• A scheduler re-runs the analysis automatically using the last saved configuration
• A failure counter system tracks consecutive non-compliant runs per store
• After a defined threshold, notifications are temporarily silenced to avoid repetitive alerts
• Email notifications are only sent when there is a meaningful state change

Deliverables

The project delivered:

• A fully functional Flask backend
• A web interface for configuration and result visualization
• Automated email summaries
• A reusable and documented backend architecture ready for extension

Part 1 - Questionnaire-Based Recommendation System (QCM)

Part 2 - Real-Time Competitive Intelligence Tool

Context & Challenge

Player transfers in top European leagues represent major financial and sporting stakes—the global transfer market generated around €6.5 billion in 2022. Clubs need to minimize the risk of costly failures (e.g., adaptation issues after a high-profile transfer) while optimizing recruitment decisions. An approach based on data is essential to assess performance before and after transfers and to identify the most relevant indicators by position and league.

Methodology (CRISP-DM)

• Domain understanding: Define the problem—how to evaluate player transfers to optimize profitability and impact on club performance.
• Data understanding: Explore available data (player performance, transfer fees, club statistics); check quality, detect anomalies (missing values, duplicates), and identify key indicators (minutes played, goals, assists, evolution after transfer).
• Data preparation: Clean and transform data with Python (Pandas): remove duplicates, handle missing values, keep only relevant indicators. Filter data for the four leagues and seasons 2017–2022; specialize analysis by position (defenders, midfielders, attackers, goalkeepers).
• Modeling: Build descriptive statistics and visualizations in Power BI (histograms, boxplots, trend and pie charts) to compare performance before and after transfer by league and club.
• Deployment: Deliver an interactive Power BI dashboard and practical recommendations for clubs (e.g., which indicators to monitor, how to reduce financial and sporting risks).
• Collaboration: Team-based project with clear role allocation, regular progress reviews, and coordinated delivery.

Dashboard & Analysis

The Power BI dashboard was designed to:

• Explore team statistics (rankings, goals scored/conceded, points) and player transfers (fees, origin/destination clubs, seasons) with dynamic filters by league and club.
• Analyze performance by position: attackers (goals, assists, goals per match), midfielders, defenders (tackles, interceptions), and goalkeepers (goals conceded, saves, distribution) before and after transfer.
• Support strategic recommendations by comparing player evolution across leagues and identifying profiles that adapted well or struggled after a transfer.

Case studies (e.g., high-profile transfers between leagues) illustrated how the dashboard can be used to assess return on investment and adaptation.

Deliverables

• Cleaned and structured dataset (Python + Pandas) for the four leagues (2017–2022)
• Interactive Power BI dashboard with KPIs and filters by position, league, and club
• Strategic recommendations to optimize recruitment and reduce financial risks
• Documented methodology and ethical considerations (GDPR, anonymization, fair-play)