Professional Summary
Senior Full Stack Developer with 13+ years architecting distributed systems and high-performance applications. Expert in TypeScript/Node.js, React, and cloud platforms (GCP/AWS/Azure). Proven track record scaling microservices to handle millions of users with sub-second latency. Strong background in observability, testing frameworks and mentoring engineering teams. Specialized in event-driven architectures and optimizing large dataset processing.
Work Experience
Meeteam — Webinar Platform · Integrations
- Role:
- Senior Full Stack Developer
- Timeline:
- February 2026 – Present (7 months)
- Team Size:
- 15+ · devs, UX, QA, PO, DevOps
Project Context: Part of the integrations team responsible for connecting enterprise clients' webinar programs (ON24/CVENT) with their external systems, CRMs, marketing automation platforms, and paid-media lead capture, powering demand-generation and lead-routing workflows across NA and EU regions.
- Designed and maintained 15+ production integrations pushing ON24/CVENT webinar registration, attendance and engagement data into client martech stacks (Salesforce, Marketo, HubSpot, Eloqua) via Tray.io embedded workflows and TypeScript/Node.js/.NET services, processing 100K+ lead records per month with 98% delivery success rate.
- Built an end-to-end LinkedIn → ON24 registration pipeline on the LinkedIn Marketing API, implementing webhook challenge-response validation, leadNotifications subscription lifecycle management with rate-limit-aware cleanup, and dynamic form-field mapping, eliminating 8 hours/week of manual CSV list uploads per client.
- Shipped self-service configuration tooling (Tray.io Config Wizard selector scripts for ad accounts, lead gen forms and audience segments) and maintained the ON24–LinkedIn segment mapping across NA/EU clients, enabling account managers to set up integrations without engineering involvement in 90% of new deployments.
- Engineered an end-to-end ON24 → Salesforce Pardot (Account Engagement) pipeline, handling Salesforce OAuth 2.0 authentication with per-client Business Unit scoping, prospect upserts through the Pardot API v5 with email-based deduplication, mapping of registration/attendance/engagement attributes to client-specific custom prospect fields, and automated list membership updates that trigger downstream nurture programs.
- Delivered an end-to-end ON24 → Microsoft Dynamics 365 integration on the Dataverse Web API, implementing Entra ID (Azure AD) app-only OAuth token management, OData upserts against leads/contacts using alternate keys to guarantee idempotency, $batch requests for bulk post-webinar syncs, and optionset/lookup resolution for per-client schema variations, syncing 10K+ records per month with automated retry handling on API throttling (429 / Retry-After).
SUMZ — Financial Services Platform
- Role:
- Lead Senior Full Stack Developer
- Timeline:
- October 2024 – February 2026 (16 months)
- Team Size:
- 5 devs · 2 UX · 1 PO
Project Context: Led development of a high-volume financial services platform processing payments, loans, and account management for BNPL clients.
- Architected cloud-native microservices platform on GCP serving 1,000+ customers by migrating Elixir/Phoenix to Node backend APIs with PostgreSQL database, deployed across Cloud Run instances with auto-scaling, achieving 99.97% uptime and sub-200ms API response times for critical endpoints.
- Implemented CI/CD pipeline using GitHub Actions and Cloud Build with automated testing (Vitest unit tests), Docker containerization, and Terraform infrastructure-as-code for GCP resources (Cloud Functions, Artifact Registry, Cloud Run), enabling 15+ production deployments per week with zero-downtime releases.
- Eliminated 90% of production incidents (from 10/month to <1/month) by implementing comprehensive observability stack using Cloud Monitoring, custom Prometheus metrics for business KPIs, and structured logging with log aggregation, reducing mean time to resolution (MTTR) from 4 hours to 45 minutes through improved debugging capabilities.
- Led migration from monolithic architecture to event-driven microservices by designing asynchronous communication patterns using message queues, implementing saga pattern for distributed transactions, and breaking down 45K-line monolith into 7 independent services (auth, payments, notifications, reporting, user-management, loan-processing, audit-logging), reducing feature deployment time from 2 weeks to 3 days.
- Established technical leadership practices by conducting weekly architecture review sessions, implementing code review standards (minimum 2 approvals, 100% test coverage requirement), creating comprehensive API documentation using OpenAPI/Swagger, and mentoring 3 junior-level developers to mid level through pair programming and design pattern workshops, resulting in 40% reduction in code review cycles.
Tranqui — Financial Agent Platform
- Role:
- Lead Senior Full Stack Developer
- Timeline:
- August 2023 – September 2024 (13 months)
- Team Size:
- IC in 12-person team
Project Context: Financial services platform enabling agents to manage customer accounts, process transactions, and handle high-volume query operations with strict performance requirements.
- Eliminated 95% of production crashes (from 15 critical failures/week to <1/week) by refactoring Node.js/Express legacy codebase, identifying and resolving memory leaks through heap profiling, implementing circuit breaker patterns for external API calls, optimizing N+1 database queries via TypeORM eager loading, and establishing request rate limiting (100 req/min per agent), saving company $45K/month in downtime costs.
- Increased test coverage from 45% to 95% by implementing comprehensive testing strategy using Jest for unit tests (2,500+ test cases), Supertest for API integration tests, and establishing CI pipeline that blocks PRs below 80% coverage threshold, reducing production bugs by 68% (from 25/month to 8/month) and improving code review efficiency through automated quality gates.
- Built real-time agent dashboard serving 200+ concurrent agents by implementing WebSocket connections using Socket.io for live transaction updates, designing React/TypeScript frontend with Material-UI components and optimized re-rendering patterns (React.memo, useMemo hooks), and implementing server-sent events (SSE) fallback for legacy browsers, achieving <300ms update latency for critical financial operations.
- Designed and deployed database migration strategy for 15M+ record PostgreSQL database by implementing zero-downtime migration approach using blue-green deployment, creating indexes online (CONCURRENTLY flag), partitioning large tables by date ranges (monthly partitions for transactions table), and establishing automated backup/restore procedures with point-in-time recovery (PITR), reducing query times on historical data by 85% (from 12 seconds to 1.8 seconds).
- Enhanced security and compliance by implementing OAuth 2.0 / JWT authentication, role-based access control (RBAC) with 15 permission levels, API request signing for sensitive operations, PCI-DSS compliant logging (masking card numbers, PII data), and automated security scanning using Snyk for dependency vulnerabilities, achieving zero security incidents during tenure and passing external SOC 2 audit.
Genius Sports — Real-Time Sports Betting
- Role:
- Senior Full Stack Developer
- Timeline:
- June 2021 – June 2023 (24 months)
- Team Size:
- Cross-functional · 8 engineers
Project Context: High-performance real-time betting platform delivering live sports widgets to major clients (NFL, NBA, Caesars, FanDuel) requiring sub-second latency and handling millions of concurrent users during peak events.
- Delivered real-time betting widgets serving 2.5M+ concurrent users during NFL Sunday games by architecting event-driven microservices architecture using React/TypeScript frontend, Node.js backend APIs, GraphQL data layer for efficient data fetching (reducing over-fetching by 60%), WebSocket streaming for live odds updates, and AWS Lambda functions for odds calculation, achieving P99 latency of 380ms and 99.95% uptime during Super Bowl traffic spikes.
- Optimized frontend bundle size by 47% (from 920KB to 487KB gzipped) by implementing advanced code-splitting strategies using React.lazy, dynamic imports for route-based splitting, tree-shaking unused dependencies, optimizing asset loading with lazy-loaded images and progressive JPEG encoding, and leveraging CDN caching (Akamai), improving Time-to-Interactive (TTI) from 4.2s to 1.8s on 3G networks and increasing user engagement by 23%.
- Built distributed caching layer reducing database load by 82% (from 12,000 to 2,200 queries/minute) by implementing Redis Cluster with consistent hashing for horizontal scalability, cache-aside pattern for frequently-accessed odds data, cache warming strategies before major events, and implementing cache invalidation using pub/sub messaging, supporting 5x traffic growth during March Madness without infrastructure scaling.
- Achieved sub-second end-to-end latency for live odds updates (<650ms from odds change to UI update) by designing high-performance data pipeline using AWS SQS for message queuing (processing 50K+ messages/second), implementing WebSocket connection pooling (supporting 100K+ concurrent connections per instance), optimizing GraphQL resolvers with DataLoader for N+1 query prevention, and using React virtualization for rendering large lists (10K+ betting options).
- Led mobile app performance optimization for React Native betting apps by implementing native module bridges for critical performance paths, optimizing list rendering with FlashList, implementing app bundle size reduction (from 35MB to 18MB through code splitting and asset compression), and designing offline-first architecture with local SQLite cache and background sync, improving app store rating from 3.8 to 4.6 stars and reducing crash rate by 78%.
- Established testing excellence by building automated test suite with 95% coverage using Jest (4,500+ unit tests), Cypress for E2E testing of critical user flows (registration, betting, withdrawal), and load testing with JMeter simulating 100K concurrent users, implementing test-driven development (TDD) practices that reduced production bugs by 71% and enabled confident continuous deployment with 30+ releases per month.
Intergrupo — Consulting · EPM Utilities
- Role:
- Junior Developer → Solution Architect
- Timeline:
- March 2013 – June 2021 (8 years 3 months)
Project Context: Promoted from Junior Developer to Solution Architect within 8 years. Led 15+ enterprise projects for EPM utilities company.
- Promoted from Junior Developer to Solution Architect within 8 years by leading 15+ enterprise projects for EPM utilities company, resulting in $2M+ annual cost savings through Azure cloud migration and process automation.
- Architected microservices platform serving 2.5M+ utility customers by designing .NET Core APIs, Azure Functions, and SQL Server backend, reducing system downtime from 12 hours/month to 2 hours/month.
- Mentored 15+ junior developers to senior level by establishing code review practices, technical design documentation standards, and conducting weekly architecture workshops.
Education
Bachelor's Degree in Systems Engineering — Institución Universitaria de Envigado
2010 — 2014
Technical Skills
- Languages: TypeScript, JavaScript, C#, Elixir
- Frontend: React, Next.js, Remix, HTML/CSS, Tailwind, MUI, Phoenix, Lit
- Backend: Node.js, Express, NestJS, .NET Core
- Databases: PostgreSQL, SQL Server, Oracle, MongoDB, Redis
- Cloud Platforms: GCP (Compute Engine, Cloud Functions, Cloud Run), AWS (Lambda, EC2, S3, SQS), Azure (App Services, Functions)
- DevOps & Infrastructure: Docker, Kubernetes, Terraform, CI/CD (GitHub Actions, Azure DevOps), Git, GitLab
- Observability: Prometheus, Grafana, DataDog, CloudWatch, Application Performance Monitoring
- Testing: Vitest, Jest, Cypress, JMeter, Unit/Integration/E2E Testing
- Message Queues: Pub/Sub, AWS SQS, RabbitMQ, Event-driven architecture
- APIs & Integration: RESTful APIs, GraphQL, WebSockets, SOAP, Webhooks
- Methodologies: Agile, Scrum, Microservices, Distributed Systems
Languages
- Spanish: Native
- English: Professional