HENDRIX ROA
Senior Software Engineer
-
Cali, Colombia-
Senior Software Engineer with over 10 years of experience in building and deploying scalable microservices.
Skilled in developing RESTful APIs, GraphQL, and WebSocket APIs using Python, Node.js, TypeScript, Nest.js,
and AWS. Proven track record in optimizing data workflows and enhancing system reliability. Experienced in
creating web scraping processes for ETL pipelines and integrating with third parties. Strong leadership in
improving code quality and DevOps practices. Proficient in backend architecture to create custom platforms
and support AI/ML-driven solutions for data analysis, automation, and decision-making.
EXPERIENCE
Senior Backend Engineer at TelevisaUnivision (2023 - Present)
● Implemented robust integrations with Quickplay’s API using Node.js and Fastify, automating a complex
media processing pipeline analogous to data ingestion and preprocessing for ML models on GCP
Vertex, reducing manual intervention by 40%.
● Integrated the Spectrum vendor into the authentication system via REST API using Nest.js, resulting in a
15% increase in user acquisition for vix.
● Architected and implemented a high-throughput video asset processing pipeline (2K to 5K assets/day)
using Nest.js and MongoDB ETL, demonstrating capabilities in designing scalable data ingestion and
transformation systems fundamental for AI/ML applications.
● Built high-performance GraphQL APIs with NestJS and Apollo, reducing client data over-fetching by 40%
and improving query response times by 30% through optimized schema design and resolver efficiency.
● To solve a critical data gap, I built a web scraping pipeline with Puppeteer to create a proprietary news
dataset. This system fed real-time information to NLP systems, creating the Knowledge Base for a
Retrieval-Augmented Generation (RAG) system used in the news feed.
● Led an initiative to improve the codebase by prioritizing unit testing and adopting TDD techniques with
frameworks such as Jest, Mocha, and Chai, resulting in a reduction of 40% production bugs.
Backend Lead & DevOps at Sonatafy (2022 - 2023)
● Designed and developed a Nest.js backend system from scratch for secure financial data integration to
ERPs (Microsoft Business Central), automating critical financial workflows and reducing report
generation time by 50%.
● Tasked with building a new FinTech platform from scratch, I architected a highly reliable, event-driven
system. I used Nest.js and Redis queues, similar to an AWS SNS/SQS architecture, to create a real-time
feature pipeline for analytics and predictive models, ensuring 99.9% uptime.
● Designed the entire data architecture for the platform, making the critical decision to use Amazon RDS
(Oracle) for transactional integrity and MongoDB for flexible, unstructured data archiving and analytics
on over 10,000 daily transactions.
● Implemented an event-driven REST API using Redis-based queues for real-time monitoring of financial
statements, improving system uptime to 99.9% and reducing processing errors by 30%, essential for
reliable financial transaction processing.
● Designed MongoDB data structure for archiving historical processing data and extracting pipeline
performance metrics, automating bi-weekly reports on 10,000+ daily transactions, showcasing skills in
data warehousing and analytics.
● Contributed to front-end tasks using React and React Native, leveraging in-depth back-end knowledge
to facilitate rapid integration with web and mobile platforms.
Backend Lead & DevOps at Auditmation (2021 - 2022)
● Leveraged Terraform to automate cloud infrastructure deployment on AWS, ensuring consistent,
scalable environments critical for deploying and managing AI/ML workloads and secure financial
applications.
● Designed and implemented GitOps CI/CD pipelines (ArgoCD, Kubernetes) and event-driven
architectures (AWS SQS, SNS, Lambda, ECS, API Gateway), enhancing system scalability and
responsiveness vital for high-volume 4 millions of transaction processing and real-time.
● Implemented AWS SSO and robust security practices (encryption, access controls, vulnerability
assessments), ensuring adherence to industry standards for secure financial data handling.
● Work on automated monitoring and reliability was so effective it reduced the team's on-call rotation
from a daily check to a single weekly review, freeing up significant engineering time.
● Confronted with a slow, manual image moderation process, I designed and deployed a computer vision
solution using AWS Rekognition. This system automatically analyzed and flagged thousands of images,
drastically reducing the team's manual workload.
Backend Lead & DevOps at DevBase (2019 - 2021)
● Led backend development for Sage Intacct Property Management: Developed Nest.js REST API for ERP
integration; engineered AWS SQS ETL for transforming 500K+ daily financial transactions (bills, invoices)
into AWS RDS, demonstrating expertise in large-scale financial data processing pipelines.
● Implemented AWS Rekognition for automated detection of advertisements and phone numbers in real
estate images, reducing on-call rotation workload by automating the initial validation of self-promotion
by real estate agents, and enabling content moderation features.
● Designed secure AWS (SFTP, S3, SQS) medical record ingestion from 10+ legacy vendors; developed
SQS-powered ETL processing 500K+ daily records into RDS PostgreSQL, highlighting experience with
complex data integration and standardization analogous to preparing diverse datasets for ML.
● Automated infrastructure (Terraform, CDK), serverless solutions (Lambda reducing costs by 35%),
reusable Lambda layers (3X deployment speed), ELK stack for monitoring, all contributing to robust,
cost-efficient, and observable systems necessary for demanding FinTech AI applications.
Backend Lead & DevOps at Supermassive (2017 - 2019)
● Developed microservices and ETL processes (Node.js, TypeScript) to collect and analyze cryptocurrency
data, identifying patterns to notify 10k subscribers of investment opportunities, showcasing early
application of data analysis for financial insights.
● Implemented the infrastructure to gather data from different crypto feed websites using playwright
and puppeteer.
● As the platform grew, I identified a scalability bottleneck and led the migration to a serverless,
IaC-based system on AWS. This created an efficient model serving infrastructure that cut deployment
time by 35% and cloud costs by 40%.
● Reduced production error rates by 35% through TypeScript adoption and API data validation in a
microservice system, critical for data integrity in financial and AI systems.
● Migrated microservices to IaC (Serverless Framework, AWS), cutting deployment time by 35% and AWS
costs by 40%, demonstrating efficiency in deploying and managing scalable cloud-native applications
suitable for AI model serving.
Backend Engineer at Strapp (2016 - 2017)
● Developed lightweight batch processing scripts with PHP Laravel. Facilitated real-time sales reporting
and improved data accuracy.
● I designed and built a data analytics module to track and analyze metrics for over 15,000 users. This
system delivered real-time insights that enabled data-driven marketing strategies and doubled sales
opportunities.
● Engineered a data processing pipeline with Node.js to automate real-time sales reporting, ensuring
accurate data analysis for stakeholders.
● Developed a secure and scalable RESTful API that served as the core data access layer for a file-sharing
application, making the architectural decision to use AWS S3 for efficient and cost-effective object
storage.
Junior Software Engineer at 4Geeks (2015 - 2016)
● Promoted from an internship, this is where I learned to translate academic knowledge into real-world
solutions. I developed foundational soft skills in collaboration, communication, and navigating team
dynamics that I've applied throughout my career.
● Built data integration pipelines using Python and Odoo to connect with ERP systems. This automated
key business processes by streamlining data flow, showing my early focus on solving problems through
data-centric automation.
● Recognizing the limitations of a legacy system, I took the initiative to lead its migration to a modern
Python (Django) architecture, improving system efficiency.
● My work on multiple client projects with diverse stacks (Laravel, Angular.js) taught me to be a polyglot
engineer, focusing on the problem rather than being tied to a single technology.
CERTIFICATIONS
●
●
●
●
●
●
●
Artificial Intelligence Foundations: Machine Learning - 2024
Agile Software Development - 2023
Choosing a Database: PostgreSQL, MySQL, MongoDB, and Cloud - 2022
Software Architecture: Domain-Driven Design - 2022
Advanced Docker Automation with AWS CDK - 2022
Node.js: Microservices - 2022
AWS Well-Architected Framework: Sustainability Pillar - 2022
SKILLS
● Languages & Frameworks: Python, Typescript, Node.js, Nest.js, TypeORM, Prisma
● Web Scraping: Playwright, Puppeteer
● API & Integration: REST APIs, GraphQL, Auth0, AWS Cognito, Firebase Auth.
● Databases: PostgreSQL, MySQL, Redis, MongoDB, AWS ECS, ECR, EC2, VPC, ALB, API Gateway, S3, RDS,
SNS, SES, Memcache, OpenSearch, Cloudwatch, EventBridge
● Cloud: AWS, GCP
● DevOps: CI/CD Pipelines, Github Actions, Terraform, AWS CDK, Pulumi, jest, mocha
