Senhaji Rhazi hamza

Context : BNP Paribas is a French international banking group, I joined their global investing arm BNP Corporate and institutional Banking (BNP CIB), BNP is building their own private internal cloud, each service now is or is about to furnish it’s value in an automated way. I joined the data agility service, as a Tech Lead to help them with their automation by building APIs and workflow around the product and services they furnish to the other teams. Mission & realizations : Our service team was based between Paris, London and Madrid, most (70%) are DBAs, the dev team I was responsible for (3-4 other devs) were in Paris. Before being completely entrusted with the Tech Lead position, the biggest technical challenge I had to take, was a complete rewrite of the main legacy software our service is relying on. Once the rewrite was successful, with the team, we made sure to stabilize the delivery of our products by building monitoring tools, improving the logs introducing smarter strategy to handle failures of the external systems our apis were interacting with. We were constantly trying to facilitate the work of our support by shortening their feedback loop to customers or adding value to our customer trough features I was also responsible, of 4 kubernetes cluster in 3 regions (Europe/middle-east, America, Asia-Pacific) with multi-cluster federation architecture. Main stacks in this environment : Python, SqlAlchemy, Ansible, Tower, Kubernetes, delphix, cyberark

Context : I worked with LVMH Paris, the first luxury brand in Europe who decided to power its subsidiaries with a cloud data-platform The aim was furnishing to the “maisons” tooling as they can instantiate a GCP cloud infrastructure corresponding to their use case, the project was labelled Atom Mission & realizations : The main tasks could be expressed as follows : Writing Terraform modules for GCP Writing connectors (BigQuery, Sale forces, GCS) Writing services REST APIs (python/flask) Technical environment : Cloud DevOps : GCP, Terraform Back-end : Python/Flask/Golang

HaroldWaste is a company that commercialize a Saas solution for the recycling market, the saas solution helps its costumer to trade waste, the solutions helps from provisioning, pricing up to transportation their product is consumed as web app and mobile app My work mainly consisted on : Developing additional features for their products on the existing stack (React, React native, GraphQl, node js, typescript, RabbitMQ) Developing independent microservices and introducing new stack (python) under their technical constraints like multi tenancy database, some of the technical constraints solved resulted this article https://medium.com/@hamza.senhajirhazi/how-to-handle-schema-multi-tennancy-with-python-flask-sqlalchemy-postgres-7000dda10749

I worked at Artefact as a data engineer, my work mainly consisted in building and designing softwares and architectures that might ship data-scientist's models (shipping them in a product), or handle big amount of data (ETL for example) Some projects worked on at Artefact : Chatbots with international beer company: - Description : I worked within a team on developing two messenger chatbots for this company, the first chatbot was build as loyalty consumer program, where a consumer take a picture of his consumption (either a bill of purchase or a special glass), this picture is analyzed by machine learning models, then once the picture is recognized, the consumer receive some credits points which gave him some privileges The second messenger chatbot was treating automatically most of the customer complains - Technologies used : - Python (sqlAlchemy + flask) - Dialogflow - Google cloud platform - PostgresQL Media dashboard for an international pharmaceutic company - Description : This company make advertisements for some of their drugs using different canal (facebook, instagram, youtube, google ads, dv360, google analytics) In order to help this company to improve the efficiency of their marketing campaign we built for them a pipeline data architecture + dashboard, that allows them to better track the performance of their advertisements campaigns - Technologies used : - Python - Amazon web service (S3) - Docker - Kubernetes (jobs scheduling)

Freelance : Mission #1 Context : Wildcax is a startup in the oil’s industry that want to revolutionize the investments on wells, since in oil industry, wells have a life cycle and it’s more interesting to invest when the wells still have much oils/gaz to produce Problem : Wildcax needed to predict the Wells's life cycle, to have models that can fit the data and predict how the wells production is going to behave Work done : based on some scientific materials : - Traditional decline analysis theory (http://www.fekete.com/san/webhelp/feketeharmony/harmony_webhelp/content/html_files/reference_material/analysis_method_theory/Traditional_Decline_Theory.htm) - JAMES CODY STATTON’s thesis (https://oaktrust.library.tamu.edu/handle/1969.1/ETD-TAMU-2012-05-11144) - Some research papers like (https://www.onepetro.org/conference-paper/SPE-134231-MS) I have build some optimization models in order to find the best parameters for the models used in the scientific resources indicated I had also to improve the convergence of my models, which led me to do some data signal processing like smoothing data or some frequency analysis using Fourier transform Techno used : -python, Scipy, pandas, matplotlib Mission #2 Context : Zeclef (http://zeclef.com/) is a startup based on Paris in the cleaning business, Zeclef have an uber model-like for cleaning Problem : -Zeclef had some problems for planing deliveries efficiently in Paris -Zeclef had also a problem for generating bills for there auto entrepreneurs Work done : -I build a solution using algorithm Traveling sales problem (https://en.wikipedia.org/wiki/Travelling_salesman_problem) but they used another SAS solution more mature that answer better to their use case I build a billing system using their excel inputs that generate all their bills, they go from 2 days each 2 weeks of work to 2 min for generating bills Techno used : -python, xlmswriter, vba, pandas

My work at reciTAL consisted in doing research on NLP (natural language processing) task, more specifically for QA systems (Question answering systems) and to develop a commercial product out of our results. In other words developing an intelligent search engine, where people can query their unstructured data in human language. QA systems task is a sub-field of NLP which is a sub-field of AI, where the challenge is to query over unstructured or semi structured data like Wikipedia articles for example, and have a system that can point the answer on the article, Stanford university have launched a competition over this task, they released a famous dataset (SQUAD dataset ) and people all over the world were competing to have come up with the best QA system During my internship, i had to do a research part in order to improve the models and adapt them for french as well as in English and develop a commercial product using those models. At that time our work at reciTAL lead us to occupy the 3rd place in one of the task (predecting the existence of answer) on the leader-board

Improving the management of the supply service. Development of a VB.net + Access application. Features: Traceability of the records within the Service. Establishment of a performance indicator related to the treatments.