~Senior Full-Stack Engineer / AI SystemsDavid Gómez

I build backend services and AI agents for a cloud-native P&C insurance platform, and run agentic tooling across the development workflow — architecture, code, review, testing.

From Venezuela to Colombia, from code to AI—always learning, always building.

Based in
Bogotá, Colombia
Remote since
2017 · U.S. companies
Current
TEAM International / BriteCore
Focus
Agentic AI · Python · Cloud
David Gómez

In progress — Building and improving the skills of AI agents for the BriteCore platform — MCP servers, deepagents

01About

Nineteen years of building, nine of them remote.

Education
Computer Engineer, 2008
University
Fermín Toro, Venezuela
Thesis
Two special distinctions
Languages
Spanish · English

I'm a Computer Engineer, originally from Venezuela and currently based in Colombia since 2019. I graduated in 2008 from Fermin Toro University in Barquisimeto, Venezuela, where my thesis earned two special distinctions—Mention for Publication and Honorific Mention—and was even featured in a local newspaper.

Since 2017, I've been working remotely with U.S.-based companies, gaining extensive experience in multicultural and distributed teams. This has shaped my ability to communicate effectively across cultures, collaborate seamlessly in remote environments, and thrive in diverse professional settings.

Throughout my career, I've built a reputation for being organized, approachable, and resilient. I enjoy tackling complex challenges, driving projects to completion, and supporting my team every step of the way. My current passion is exploring the world of artificial intelligence, while continuing to grow as a leader and mentor.

02Experience

Where I have built things, and what came of it.

Apr 2020 — Present · United States · Remote

Cloud-native, multi-tenant platform for P&C insurers. I build the conversational AI agent layer — the agents themselves, the MCP servers they call, and the persistence, authorization, and safety work that makes them production-grade — and apply agentic tooling across the development workflow.

  • Python
  • FastAPI
  • Flask
  • LangGraph
  • MCP Servers
  • Agentic AI
  • Vue.js
  • MySQL
  • Alembic
  • Docker
  • Kubernetes
  • AWS
  • GitHub Actions
  • k6
  • Django

Key achievements

  • Build conversational AI agents for a multi-tenant P&C insurance platform, including LangGraph deep agents with custom middleware that bounds tool retries and ends a run with an actionable message instead of crashing on the recursion limit
  • Design and build the MCP servers those agents call, turning policy, lines, quoting, claims, rules, and document APIs into governed agent tools
  • Build the persistence layer behind agent jobs and chat: MySQL-backed with an Alembic-owned schema, long-running jobs resumable after a pod restart, and per-user isolation enforced at both the route and storage layers so a missed ownership check cannot be exploited
  • Propagate caller identity from agent through MCP to the core API and audit every tool call, making agent-initiated changes attributable and turning opaque permission failures into actionable messages
  • Harden agent output against disclosure — keeping reasoning traces and internal file paths out of user-facing chat, and training agents to refuse unsupported mid-session input
  • Add human-in-the-loop approval gates and streaming progress to long-running agent workflows, so users review and correct extracted data before anything is committed
  • Roll the agent stack out across carrier environments, pinning builds and tuning per-tenant deployment configuration
  • Design backend services in Python, FastAPI, and Flask; completed the platform's Python 3 migration and architected a policy numbering service using optimistic locking for safe concurrent issuance
  • Introduce k6 load testing, including concurrent policy issuance stress tests, and remediate security vulnerabilities such as HTTP parameter pollution in API gateway routing
  • Review 370+ pull requests across the platform and agent repositories, and mentor engineers through architecture discussion and code review
  • Apply agentic coding tools (Claude Code, Cursor, GitHub Copilot) across the development lifecycle for implementation, review, documentation, test generation, and story creation
Company site

Sep 2017 — Dec 2019 · Venezuela

Key contributor to the startup's rapid product development process. Led the team in implementing designs using the latest web development technologies, and managed Public Cloud Services (GCP).

Nov 2011 — Aug 2017 · Barquisimeto, Venezuela

Startup owner. Designed and implemented a scalable, high-performance tech network to get the business operational quickly, and led the team through business administration and strategic planning.

Dec 2009 — Jul 2010 · Barquisimeto, Venezuela

Collaborated on the non-bank correspondents project for Casa Propia E.A.P., focusing on presentation layers and middleware applications using Crosswalk language.

Aug 2008 — Nov 2009 · Barquisimeto, Venezuela

Developed a Visual Basic-based system for ordering menus in a restaurant, and participated in a system testing team focused on exception and error routines, validations, and limit values.

Jan 2008 — Jul 2008 · Barquisimeto, Venezuela

Designed and developed a Visual Basic-based system to control the weight of chemicals in custom formulas, integrating both software and hardware components. The application is still in use.

Oct 2007 — Dec 2007 · Barquisimeto, Venezuela

Completed Computer Engineering internships in the Lara Transmissions department, gaining extensive knowledge of telecommunications and data transmission systems.

Apr 2007 — Jul 2007 · Barquisimeto, Venezuela

Designed and developed a Visual Basic-based system for controlling and printing labels for tire storage, integrating software and hardware to improve inventory management.

03Capabilities

What I build, and what I build it with.

Backend systems and AI agents first, with the frontend and platform work to ship them end to end.

A AI & Agents

  • Agentic AI workflows

    Claude Code, Cursor, Copilot across coding, review, docs, tests, story creation.

  • MCP servers

    Production servers letting AI agents drive domain APIs.

  • AI product features

    Claims analytics agent with feature flags and streaming support.

  • Chat interfaces

    Vue.js front ends for AI-powered features.

B Backend

  • Python

    Scalable server-side applications.

  • FastAPI

    Layered services for multi-tenant platforms.

  • Flask

    APIs and internal services.

  • Django

    Full-featured application backends.

C Frontend

  • JavaScript / TypeScript

    Dynamic, interactive web applications.

  • Vue.js

    Modern, responsive interfaces.

  • React.js

    Component-driven user interfaces.

  • Next.js

    Server-rendered React applications.

D Platform

  • AWS

    High-availability, scalable deployments.

  • Docker & Kubernetes

    Containerized apps and local dev environments.

  • GitHub Actions

    CI/CD pipelines.

  • k6

    Load testing, including concurrent-issuance stress tests.

04Contact

Let's work together

Open to conversations about AI-augmented engineering, platform work, and senior full-stack roles.

Location
Bogotá, Colombia
Time zone
GMT-5 · U.S. hours