Filipe A.

Viridios AI (VAI) develops an analytical platform for the carbon market, providing prices and analytics insights for multiple carbon assets using in-house AI and quantitative models. As a Quant Data Scientist, I’m responsible for developing and optimizing the existing pricing models.

Model Research: Developed and Implemented a time series pricing model that was used to bring old market quotes into present value which was used to improve the existent pricing methodology;

Automation: Developed and implemented automation pipelines for extracting, validating, processing, and publishing daily carbon asset prices, reducing process time by 75%;

Enhancing existing models: Developed and implemented a new methodology for estimating specific idiosyncratic factors of carbon assets of the Voluntary Carbon Market (VCM), the improvements added more granularity providing better estimation of factors for specific carbon credit groups;

Cross-functional leadership: Led the expansion of the pricing methodology to the Australian Compliance Market, collaborating with Data Engineering, Software Engineering, and Data Operation teams;

Monitoring: Developed and implemented automated model and data validation pipelines validating if previous patterns and model behaviors are still observed in the present, and also checking for human error in an early stage of the process. Reduced the process reiteration due to human error by 85%.

BTG Pactual is the largest investment bank in Latin America, managing assets exceeding R$1.5 trillion. As a Data Analyst, I was embedded within the Digital Retail Unit. My role involved bridging the business and analytics teams, developing dashboards, and generating key insights to support business decisions.

Dashboard development: Designed and implemented dashboards using Power BI, SQL, Python, and Spark to track daily and monthly KPIs. These dashboards were critical in measuring banker performance, team leader efficiency, and overall business health, achieving 80% daily usage and 100% usage every two days by key stakeholders.

ETL & data ingestion: Used AWS and PySpark to automate data extraction, transformation, and ingestion (ETL) processes, optimizing data flow from various sources into the reporting systems. This significantly reduced manual efforts and ensured up-to-date, accurate data for decision-making.

Process automation & data quality: Developed a quality process for data validation using Apache Airflow, automating data validation before data updates in the dashboards. This reduced potential errors and ensured seamless, accurate data pipeline updates, minimizing disruptions in reporting. Reduced 90% the requested time to update and validate dashboard results before publishing to the business teams;

Training & Collaboration: Developed a training program for incoming analysts, ensuring a smooth transition and knowledge sharing. The program was well-received and praised for its clarity and practical approach, ensuring the new team members were well-prepared;

Siemens is the largest industrial manufacturing company in Europe and is a global market leader in industrial automation and software. It is focused on industrial automation, distributed energy resources, rail transport, and health technology. Worked in the innovation lab, prototyping applications that would be used in Siemens internal teams.

Worked with genetic algorithms in the prototype of an app to automatically update office layouts to optimal parameters like distance between tables, and natural and artificial light;

Worked with the Internet of Things and mobile development in the prototyping of mobile apps;