Gustavo Sosa Cuervo

About Gustavo Sosa Cuervo

After years of working deep in backend systems, I’ve learned that most engineering problems aren’t caused by lack of talent or effort. They come from systems that outgrow their original assumptions. Data volumes change, product direction shifts, and suddenly the architecture that once felt clean starts working against the teams relying on it. That’s the space I’ve spent my career navigating.I’m a senior software engineer with 8+ years of experience building and reshaping distributed systems in data-heavy environments where reliability is non-negotiable. My work lives behind the scenes, focused on how information moves, how services interact, and how platforms behave under real operational pressure. I’m drawn to complex systems not to make them clever, but to make them understandable, resilient, and easier to evolve without breaking trust.Much of my role has involved stepping into platforms that are already in motion, often strained by scale, and helping restore structure to them. I work closely with product, analytics, and engineering partners to turn ambiguous requirements into clear technical direction, aligning architecture with how the business operates today, not how it once did. I care deeply about long-term system health because I’ve seen how quickly small decisions compound at scale.What motivates me is building foundations that other engineers can rely on without second-guessing. Clean APIs, predictable data behavior, and systems that fail in understandable ways matter more to me than shipping something fast that becomes tomorrow’s constraint. When the foundation is right, teams move faster naturally.A few examples of the kind of impact I focus on:• Unified 10 disconnected recommendation systems into a single platform by redesigning schemas and APIs, increasing monthly recommendation output by 20x while restoring visibility across product lines• Replaced a fragile batch workflow with an event-driven, serverless architecture handling over 100K large ML jobs per month, slashing delivery cycles by 20% and establishing a reusable pattern adopted across multiple teams• Designed a partition-aware deduplication engine processing billions of records, reducing analytics query times from hours to minutes and enabling time-sensitive decisions across science and business teamsI build for scale, but I design for clarity.If you are building or evolving systems where reliability, data integrity, and long-term maintainability matter, I’m always open to thoughtful conversations with engineers, product leaders, and teams facing those challenges.

Experience

1. Software Engineer Ii

   Amazon

   Oct 2022 — Present · Chicago, IL, US

   Delivered a campaign detection engine in days by converting noisy performance signals into production-safe logic, recovering ~$30M in at-risk revenue while giving sales teams early visibility into failing ads.* Launched an end-to-end recommendation engine spanning data models, APIs, and internal workflows, enabling advertiser-specific suggestions and driving a 10x increase in downstream ad purchases across the funnel.* Orchestrated automated ingestion and normalization pipelines processing 10GB of advertiser metrics per cycle, removing manual pulls and expanding client engagement opportunities fivefold through fresher insights.* Optimized multi-terabyte analytics workflows into fast-loading QuickSight dashboards, consolidating fragmented metrics into a daily operational command center for project managers and analysts.

Education

• University of Havana

  Bachelor's degree, Computer Science

  2011 — 2016

Skills

• Leadership
• Microsoft Word
• Public Speaking
• Microsoft Office
• Html
• Asp.net Web Api
• Powerpoint
• Microsoft Excel
• Html5
• Sqlite
• .net Framework
• C#
• Django
• Symfony
• Python
• Customer Service
• Mysql
• Research
• Php
• Cascading Style Sheets (Css)
• Microsoft Sql Server
• Javascript
• Umbraco
• Asp.net Mvc
• Asp.net
• Social Media
• Marketing