Hands-on data engineer with a proven record of deploying scalable distributed applications, and supporting data science forecasting applications in production. Experienced in a variety of engineering technologies and model evaluation techniques.
Automating the process of identifying faulty tires and alerting the client to submit a new work order from Infor EAM
Develop and manage ETL workflow using Infor ION Workflow
Working on the integration of Smart Asset Management (SAM) with the client's on-premise Infor EAM system
Refactoring the codebase to keep up with the latest reference architecture and reduce technical debt
Coordinating with solution architects to solidify the data flows between Infor CloudSuite products (SAM, EAM, Birst, Data Lake, etc.)
Working closely with data scientists, business analysts and product owners on data-related technical issues and their data needs
Lead the delivery and integration of the Smart Forecasting Management solution with the client's Infor CloudSuite tenant
Implemented a parallel architecture to run multiple Spark jobs on a partitioned data set for better ML predictions
Collaborated with solution architects from other Infor teams to integrate Data Fabric platform functionalities into the final solution
Improved the logging and monitoring of the solution and configured performance degradation and failures alerts
Wrote technical documentation on the architecture and data transformations, and trained the data scientists to use it for ML tuning
Put in place coding practices and deployment processes for the team to streamline the development and release cycles
Lead the delivery and integration of the Pricing Intelligence solution with the client's Infor CloudSuite tenant.
Put in place metrics and processes for validating the quality of the inbound and outbound data.
Contributed to the development road map of Infor Coleman by writing feature requests and bug reports to speed up the solution delivery.
Collaborated with multiple stakeholders like data scientists, business analysts and product owners to quickly solve technical issues and ensure good communication with the clients.
Wrote and reviewed technical documentation, data interface specifications, and extension guides.
Maintained a platform for running Spark applications on managed cloud services, mainly on AWS
Developed Spark applications to make the bootstrapping of Infor products faster and repeatable. The applications cover data transformation, schema and constraints validation, outlier detection and ML. The time-to-delivery dropped from 15-20 business days to 2-4, and the forecast tuning from 5-10 business days to several hours
Designed, scheduled and monitored data pipelines using Airflow on AWS MWAA
Established coding best practices (pair code reviews, code linting configuration, git hooks, etc.)
Adopted Semantic Versioning as our versioning policy for the releases to reduce the risk of breaking software for existing clients
Wrote technical documentation on the internals of the system and how-to guides to make the onboarding of engineers easy
Setup CI/CD pipelines to build, test, package and publish each artifact
Drove the technical discussions with the client around integration and data interface specification of Demand Management
Assisted in the Exploratory Data Analysis by running a data profiling pipeline to identify patterns and get metrics and statistics.
Developed SQL transformations to map the client's data specs to be ingested into Demand Management.
Performed end-to-end testing of the solution: transformation of the data, ingestion into Demand Management, and generation of the ML forecast.
Designed a forecast assessment pipeline to evaluate the previous forecast outputs of the ML engine against the incoming data containing the actual sales.
Integrated the Demand Management solution with the Infor SCP to automate the creation and review of work orders.
Prepared a workspace in Infor Birst to visualize the results of the EDA and the generated ML forecast.
Took part in technical discussions and worked closely with the Demand Management development and support teams.
Validated the inbound client data against a list of constraints (primary keys, completeness, datatypes, etc.) and ran data profiling to complement Exploratory Data Analysis
Performed end-to-end testing of the solution: ingestion into Demand Management, and generation of Time Series and ML forecasts
Monitored the workflows and optimized the predicates queries for better performance to meet SLAs
Wrote technical documentation covering the high-level architecture of the solution, integration with external systems, batch operations, and data interface specifications
Setup integration with the client on-premise SQL Server to pull data from relational tables.
Developed data transformations to map the client data into the data specification of Demand Management.
Performed end-to-end testing of the solution: from data transformation, ingestion into Demand Management, and generation of the final ML forecast.
Monitored the workflows and optimized the predicates queries for better performance to meet SLAs.
Wrote technical documentation covering the high-level architecture of the solution, integration with external systems, batch operations, and data interface specifications.
Responsible for the build and packaging of the Demand Management Nix-based image using the LogicBlox platform
Designed data models and hierarchy structures using LogiQL predicates in the LogicBlox OLTP database
Improved system and batch performance by tuning the application design and optimizing queries
Implemented workflows using the Logicblox workflow DSL for bootstrapping the application, ingesting data, and generating Time Series forecast, ML predictions, and replenishment reports
Managed multiple clusters for different releases: dev, UAT, predeployed, and production
Worked with multiple teams and stakeholders: Cloud OPS, DevOPS, Application Support, and Business Consultancy
Ashwin Mishra
Artificial Intelligence Consultant @ Google
"Hedi initially reported to me as an intern and eventually as a full-time data engineer. During his time, he excelled at putting together a CNN based multi-label classifier using DeepFashion. Hedi's all-round skills stood out to me as not only was he able to put together an efficient solution, he packaged it well with a web-application and a live demo which resonated strongly with executive audiences.
Hedi is calm and methodical in his problem solving and performs really well under pressure. Since our collaboration he has continued to evolve as a strong data engineering practitioner and that speaks to his commitment to continually improve.
I have no hesitation in recommending him for any future opportunities."
Firas Aloui
Data Engineering and Architecture Manager @ Infor
"Briefly put, Hedi is one of the most reliable, resourceful and efficient engineers that I worked with. He was able to exceed the expectation at multiple occasions and always delivers on his commitments.
I will always trust Hedi with critical missions!
Cheers."
Varun Sridharan
Senior Technical Product Manager @ Amazon
"Eddy is one of the top notch data engineer I had the privilege to lead when stakes were high. He came in as highly talented fresh graduate that he stunned the management with his meticulous execution skills. He showed ownership from day one by crossing boundaries to bring life to a complex demand management product. His secret sauce lies in providing solid incremental progress that he mastered this skill very quickly. These traits helped him earn well deserved reputation among his peers and leadership team. This made him the most wanted engineer not only when stakes were high but also to rescue troubled project(s).
I take pride in trusting his ability and investing in his data engineering skills. Eddy is one of the handful individuals that I would go great lengths to have in my team. Wishing him many more success now and forever!"
Call Me On
+21652456619
Email Me At
contact@hedibejaoui.me
2022 © Hedi Bejaoui.