Designing and building a Data Management platform (ArtifactDB), based on AWS services to support data management and migration efforts of genomics datasets to the cloud: microservice APIs supporting heterogenous downstream applications (R/python analytical environments, webapps, MLOps platforms...), improving access, reliability, resilience, concurrency and response times. Automatically provisioned and deployed, to provide a self-service access to the platform and promote “data product” principles inspired by Data Mesh distributed architecture. Technical environment: a lot of AWS services (S3, Opensearch, EKS, IAM, RDS, DynamoDB, SNS, SQS, etc...), secured REST API in python deployed on top of Kubernetes using Helm, CI/CD and IaC based on Giltab pipelines, Flux and Terraform.

Designed, implemented, deployed and maintained publicly accessible and updated Biothings API web services dedicated to biological research (10 to 100 million hits/month, 100 million to 2 billion documents: mygene.info, myvariant.info and mychem.info):

• Normalized, refactored and optimized existing code base to bring it to a production-ready state. Delivered easy-to-use and -maintain backend systems. Implemented efficient parallelization improving processing times from weeks to hour and allowing steady schedule updates with minimal intervention and increasing data reliability.

• Automated front-end deployment on AWS cloud (server deployment orchestration).

Architected and implemented products aimed at bioinformatics developers to easily contribute to existing Biothings APIs, build their own APIs and host them on the cloud (Biothings Studio, Biothings Farm)

• Designed and implemented a development environment and a web application, packaged using Docker, to facilitate data contribution. Reduced manual and complex tasks by creating a module auto-generating indexing rules and performing sanity checks by inspecting data. Increased number of contributions received and collaborations initiated. Continuous Integration (CI) with automated testing (unit tests, API tests, UI tests).

• Devised a hosting infrastructure and architecture, allowing developers to create and host their own API web services. Automatically deployed (system, network, security) on top of AWS services.

Machine Learning: Evaluated and prototyped time series analysis based on ARIMA and RNN LSTM in order to automatically forecast evolution of the number of documents and other metrics, and track potential anomalies in datasets produced during API data releases.

In a Technical Pre-Sales capacity, worked closely with several utilities (Gaz de France, Electricité de France…):

• Wrote responses to RFPs: based on company’s existing technological assets, designed technical and functional specifications. Explained the solution to the stakeholders during RFP presentations.

• Elaborated and evaluated detailed pricing elements (electronic components, mechanical parts, logistics, backend servers, non-recurring engineering costs…) applying the company’s margin expectations.

As a Technical Project Manager, in charge of Gaz de France’s EnergyBox OEM product, a radio WMBus/IP gateway used to collect electricity and gas consumption, operated using iOS/Android smartphones:

• Scheduled the different project phases, manage prioritization and trade-offs.

• Acted as the main technical interface for customer and other contractors. Architected interfaces between company and customer systems. Quickly developed prototypes and adapted existing code (backend servers, embedded software, iOS & Android apps, development tools).

In a privately owned asset manager, managing 6 billion euros.

Data management:

• Optimized databases (queries, stored procedures, table structures…) resulting in faster processes and reporting generation. Automated critical processes avoiding manual tasks and reducing operational risks.

• Defined operational plan to ensure date is properly integrated at expected time. Developed several tools to monitor and pro-actively detect anomalies for better data consistency and reduced maintenance.

Migration to private and public clouds:

• Defined processes to store critical data in clouds. Virtualized application servers extending computing and storage resources.

• Migrated Outlook servers to Google Apps: proof of concept, planning, migration procedures, user training and support). Developed cloud-hosted application to extend Apps features.

External providers management:

• Managed providers in charge of outsourced network infrastructure: contract, outsourced perimeter and SLAs, controlled expecting services.

In a financial broker company, in charge of in-house software development.

Financial data management:

• Designed and developed several applications to integrate and export data events in real-time, regarding front-office to back-office applications (orders, executions, bookings, settlement instructions, accounts...), building a STP (Straight Through Processing) automated pipeline.

• Developed clients and servers implementing FIX protocol, to easily interface existing third party systems.

Business intelligence and reporting:

• Gathered user requirements, designed and built web-services based software architectures, with user-oriented tools providing help to be independent while accessing data.

• Orchestrated information system data to build maintainable analytical tools for all company’s units: front-to-back, compliance, accounts, C-levels, customers.