● Architected and delivered a fully automated infrastructure for Sisense - Amazon QuickSight migration, building a robust suite of Python/FastAPI microservices to manage tenant and product provisioning, embed URL generation, permission enforcement, asset bundle deployment, and data source registration. ● Delivered a web portal for onboarding and content workflows, plus an internal end-to-end validation harness (“dogfooding”) that checks permissions, embedding, and content readiness ● Developed Databricks jobs to extract and transform resource usage and operational telemetry, integrating REST APIs and system tables to capture platform-wide metrics. Leveraged dbt to orchestrate and automate the transformation logic. ● Implemented incremental CDC from MongoDB to Databricks using Spark Structured Streaming with Delta Lake idempotent upserts, checkpointing, and catalog governance. Tech Stack: Python, FastAPI, AWS S3, AWS Redshift, Typescript, Angular, React, Databricks, MongoDB, dbt, Node.js, Express.js, JavaScript, RESTful APIs, JWT Authentication, OAuth 2.0, SQL, GitHub, Scrum, Jira.

● Engineered a SQL Server Data Warehouse with automated ingestion via Azure Functions. ● Designed APIs to aggregate sensor data and applied fan-in/fan-out patterns to maximize throughput and minimize latency for BI workloads across departments. ● Orchestrated the migration to Azure Cloud, reducing infrastructure costs by ~35% while improving performance, resilience, and scalability for future growth. ● Designed and implemented scalable ETL pipelines to ingest raw data from Amazon S3, perform transformation logic including joins, aggregations, and validations, and load curated datasets into Amazon Redshift for downstream analytics. ● Built efficient AWS Glue jobs for data cleansing and enrichment, leveraging dynamic frames and partitioning strategies to handle large volumes with optimal performance. Tech Stack: Node.js, Express.js, JavaScript, Python, Jupyter, Azure Functions, Azure Cloud, Azure DevOps, RESTful APIs, AWS S3, AWS GLUE, AWS STEP FUNCTIONS, JWT Authentication, OAuth 2.0, SQL, GitHub, Scrum, Jira.

● Designed and implemented data migration strategies to move data from Oracle to Redshift and S3. ● Coordinated and monitored data loading processes, ensuring efficient and timely delivery of data. ● Automated the deployment and maintenance of infrastructure, reducing downtime and improving efficiency. ● Resolved bugs in the data migration process, ensuring data accuracy and completeness. ● Worked on the continuous integration and deployment of data migration pipelines, improving the overall data migration process. Tech Stack: Python 3.6, AWS Step Functions, AWS S3, AWS Lambda, AWS Redshift, AWS Glue, AWS CloudFormation, AWS Cloudwatch, Oracle Databases, SQL, Github, Kanban, Azure Devops.

● Implemented several neural networks architectures (convolutional and recurrents) for detection, segmentation and extracting text from receipt recognition. So that automatization over huge amounts of images can be feasible. ● Collaborated in the development of a ML platform using differentes services from AWS to detect objects in a videos uploaded by the user. So that, user can make queries about a particular object if it appears in the video. ● Designed and Implemented algorithms to segment customers in order to increase customer lifetime value (CLV) KPI. ● Used Data Mining algorithms to process comments of customers, which were obtained after AWS Transcribe converted audio into text. This way, we could identify some patterns and correlations present in the data. ● Created an automation process to process CVs from applicants and to find best candidates for different positions using HubSpot, Zapier and AWS Lambda. Tech Stack: Python 3.6, Scikit-learn, Tensorflow, Pytorch, Matplotlib, Seaborn, SQL, Amazon Web Services, Amazon S3, AWS EC2, AWS EMR, Amazon Athena, AWS Lambda, AWS Glue, AWS Sagemaker, Serverless Framework, Microsoft Excel.

● Built and operated AWS data pipelines (S3, Glue, Athena, Redshift, Lambda) for the Data Lake & Analytics program, enabling reliable ingestion and performant querying. ● Administered and supported core analytics platforms—Trino/Starburst, Airflow, Qubole, Querybook, and JupyterHub. ● Provisioned and deployed on-demand analytics clusters (e.g., Trino/Starburst, Airflow) for partner teams, accelerating delivery of new data workloads. Tech Stack: Python, AWS services, Trino, Starburst, Airflow, Qubole, Querybook, Spark, Rundeck, Terraform, Github, Jira.

● Built demand-forecasting models to predict product sales across 40 stores, achieving ~10% mAPE consistently. ● Developed per-store workforce optimization models, delivering <5-month payback through improved scheduling and headcount allocation. ● Analyzed 40M+ records in BigQuery using SQL and stored procedures; produced store- and department-level heatmaps to identify peak/off-peak hours. ● Partnered with technical and non-technical stakeholders to surface data-driven opportunities and new features Tech Stack: Python 3.6, Scikit-learn, Matplotlib, Seaborn, Google Big Query, SQL, fb prophet, Jupyter-notebooks in GCP, Vertex AI.

● Created a new analytical tool to support similarity queries interactively on a large multidimensional dataset georeferenced. ● Proposed a new data structure based on trees in the state of art, that allows interactive similarity queries (Real-Time) of large volumes of georeferenced and mixed multidimensional (30+ numerical and categorical attributes). ● Implemented backend (C++, Python) and frontend (Javascript, Leaflet.js) sides of the new analytical tool. Besides, the visualization components allow us the exploratory analysis of the data stored in the data structures, exploiting the capabilities of similarity queries. ● Verified the usefulness of this analytical tool with multiple case studies, using real-world data. Tech Stack: C++ 14, javascript, leaflet.js, d3.js, html, Css , python, boost c++ libraries, matplotlib, numpy, pandas, scikit-learn.

● Built an advanced mobile fitness application using Firebase, Android Studio, and Java, personalizing the user experience based on individual goals and progress. ● Implemented a nutrition feature, providing personalized dietary recommendations. User data and dietary information were stored in Firebase's Firestore for real-time updates. ● Developed an interactive exercise planning tool, utilizing data analytics and Machine Learning to deliver personalized workout plans, improving user satisfaction and goal attainment. Tech Stack: Firebase, Android Studio, Java, Machine Learning, Firebase Firestore, Firebase Cloud Messaging, Firebase ML, Github, JUnit.