Imad El Achiri

- Enhanced the latency of Netsuite RAG system to generate LLM responses to users up to 50 times faster without compromising the quality of responses. -Constructed different versions of the RAG to benchmark them. -Built question/answer dataset from Oracle community discussions using scraping then answer filtering to match the right answer to the thread question. -Tested the LLM inside the rag, by building virtual agents that send concurrent requests to the LLM, to analyse its behaviour and limits.

Did multiple time series analysis on the field of finance, and studied the causality between them, then built VAR models for forecasting.

-Collected data through Digimind social platform in order to build a classification model based on industrial classes. - Used logistic regression, Naïve bayes, Linear svc and deep learning to build multiple models and benchmark them while also adapting the data. - Led to 2x faster report generation for the consulting team.