Felipe J. Canales
LinkedIn | 718-407-9033 | FelipeJCanales@outlook.com | felipejcanales.com | GitHub
Skills ____________________________________________________________________________________________
• Python | Java | JavaScript | C++ | C | MongoDB | MSSQL | Node | Maven | Quarkus | Kotlin | Kubernetes | Hadoop | Spark | Tensorflow | Keras | Tableau
• AWS | Machine Learning | CI/CD | Agile Development | Jest | Cucumber | Cloud Computing | Unit Testing | Lambda | OOP
Experience _______________________________________________________________________________________
Software Engineer II
American Express
Remote
01/2022 - Current
• Assisted in creating a downstream notification application for transactions in India (Kotlin, Quarkus, Gradle, Kafka, Docker, PostgreSQL)
- Created new API endpoints to receive downstream application response.
- Integrated internal transaction notification clients.
- Leveraged Kafka to achieve asynchrony.
- Data stored complied with India Government laws.
• Developed features such as real-time transaction monitoring, alerting, and risk scoring.
• Assisted in creating and validating an auto-retry mechanism for failed transaction for anti-money laundering team
- When a transaction fails to post to an internal system, retry posting three times automatically (previously only manual retries).
- If after three retries, post still fails, alert would be raised and a manual investigation would be initiated.
• Created no data check feature for Matching Engine application (Kotlin, Quarkus, Gradle, Kafka, PostgreSQL)
- Checker that detects if application received data from bank for wire transactions for previous business day.
- Created alert to notify team if no data found.
• Assisted in expanding application coverage from only one market to three.
- Created API endpoint to receive user account information from EMEA markets in compliance with country laws.
- Updated PostgreSQL database to accommodate changes made for new markets.
- Created feature to read paginated transactions and execute matching function for each transaction
- Created feature to grab user account number from any freeform text in transaction details
• Led meetings with supported technical teams as well as business teams.
- Identified reporting requirements
- Identified transaction formatting for new markets
- Discussed potential issues/differences with currently supported markets
- Developed new application architecture to support all incoming markets
• Increased test coverage for application from roughly 5% to 80% using Junit and Mockito.
• Participated in code reviews, paired programming, and knowledge sharing sessions to maintain code quality.
Software Engineer
M&T Bank
Remote
06/2021 - 01/2022
• Assisted in creating a cicd pipeline project template package used throughout the bank (Java Springboot, Maven, Docker, YML)
- Created user portal to output a Java Springboot template application that passed all necessary requirements.
- Leveraged RESTful API to automate policy scan of CI/CD pipelines.
• Resolved tickets involving Gitlab runners issues, pipeline issues, and more.
• Assisted in creating onboarding documentation for incoming employees who are beginners in DevOps.
• Gathered metrics to improve team’s productivity, application maintenance and timeline management, and to show stakeholders our
sprint-to-sprint improvements.
• Assisted in changing the infrastructure of the company to simplify application maintenance, creation, and support.
Machine Learning/Data Science Intern
Linde plc
Buffalo, NY / Remote
01/2020 - 01/2021
• Identified loss of revenue with rental equipment resulting in $100k increase in revenue (Python, SQL)
- Found that several renters were not being charged the monthly equipment rental fee.
- Created a script that allowed company to identify which renters were not being charged and how long they weren’t charged for.
• Maintained Equipment Condition Monitoring program resulting in a 20% reduction in downtime(Python, Tableau, Historian Database)
- Updated machine learning model used to monitor machines
- Previously only one type of machine was monitored, expanded to multiple types
- Formatted, cleaned, and filtered machine sensory data using Python and Historian Database
- Analyzed and visualized data to predict abnormalities in machines
• Analyzed and visualized company’s quoting conversion rate (SQL, Tableau) - Won an internal award
- Visualization of key quoting conversion data for stakeholders
- Led to new policies being implemented in regions with a low conversion rate
Education ________________________________________________________________________________________
University at Buffalo, The State University of New York
Master of Science in Computer Science 12/2021
