Saïba SAWADOGO

Achievements: ✅ Optimising the performance of AI models by labelling datasets (text, images, audio, etc.) ✅ Guiding AI models by writing and optimising prompts. ✅ Environments : Python, LLMs, Generative AI, Prompt Engineering

Context : Linking primary care with the emergency department, Optimising stays in the emergency department through the provision of care (SNP). Achievements : ✅ Optimized yearly data extraction and indicators building by 20% by creating a reusable and automated code template using the SNDS databases. ✅ Helped in detecting high activity physicians and increased baseline clustering models (K-Means, HCA, DBSCAN) performance by 20% by integrating features engineering and domain specific knowledge. ✅ Increased indicators accuracy and specificity by designing algorithms to detect specific pathologies based on drugs prescriptions and reimbursements. ✅ Environments : Python, R, SAS, SQL, Oracle, SNDS, PMSI, Machine Learning, Programming, Big Data

Context : Multicenter retrospective study to assess the impact of COVID-19 on urgent digestive surgery. Achievements : ✅ Processed large scale databases from 39 hospitals using Big Data optimized and distributed tools like Spark. ✅ Increased statistical and Machine Learning models performance, intended to assess mortality risk factors using feature engineering and domain specific knowledge. ✅ Participated in results dissemination through scientific communication (Co-writing a scientific article published in the World Journal of Emergency Surgery) ✅ Environments : Python, R, Spark, PostgreSQL, EDS, Machine Learning, Big Data, OMOP CDM, RWD

Context: School support Achievments : ✅ Helping students struggling with math; ✅ Liaising between Complétude and families ; ✅ Helping students regain confidence in math.

Context: Development and deployment of interactive tools (R-Shiny) to facilitate activity reporting. Achievements : ✅ Increased data quality through multisource data processing and harmonization. ✅ Increased company results dissemination by designing 2 shiny interactive dashboards, deployed for end users. ✅ Assessed COVID-19 impact on activity by implementing time series forecasting using ARIMA models. ✅ Environments : Python, R, R-Shiny, Html/CSS, GitHub, Git, Predictive analysis, Time series modeling, ARIMA modeling