Skip to content

Oscar Bucio/ Backend & AI Engineer

Backends for production AI.

I design typed services, agent tools and cloud workflows that move AI from demos into regulated products.

View workEmail me
Portrait of Oscar Bucio
Current focus
Conversational AI for bankingSoftware Engineer at Plexus Tech, BBVA project.
Experience
Backend to agent systemsFintech, API quality and process automation since 2022.
Work status
U.S. work authorizedBased in Mexico City. No sponsorship required.

Systems, shown in context.

Banking integration, financial retrieval and API quality, with the boundaries and tools that shaped each build.

Plexus Tech, BBVA project

Conversational banking integration

A typed agent boundary for banking.

Building a Python integration that connects OpenAI agents to banking services through typed tools and enterprise authentication.

  • Python
  • OpenAI Agents SDK
  • Pydantic
  • AWS
  • Enterprise auth
Typed banking integration
  1. Enterprise authenticationAccess boundary
  2. Agent orchestrationOpenAI Agents SDK
  3. Typed Python toolsPydantic contracts
  4. Banking servicesBanking integration

FIXAT

Serverless financial RAG

Financial RAG on an event-driven backbone.

Built a serverless financial RAG system with Amazon Bedrock and LangChain, supported by AWS orchestration and MongoDB.

  • Amazon Bedrock
  • LangChain
  • AWS Lambda
  • Step Functions
  • SQS
  • EventBridge
  • MongoDB
Retrieval and event orchestration

Retrieval runtime

Amazon Bedrock + LangChain

Serverless compute

AWS Lambda

Workflow orchestration

Step Functions

Messaging

SQS

Event routing

EventBridge

Persistence

MongoDB

Hitss México, Claro Drive

API quality

Repeatable API quality for Claro Drive.

Tested Claro Drive APIs with Postman suites and automated Python and JavaScript checks.

  • Postman
  • Python
  • JavaScript
Claro Drive API verification

Claro Drive APIs

System under test

Verification layer

Postman suites

Python checks

JavaScript checks

Engineering that stays understandable under pressure.

I keep AI systems grounded in explicit interfaces, controlled access and workflows that teams can operate.

Architectural layers connected by controlled blue signal paths

Contracts before cleverness

Typed boundaries keep business rules, agent tools and infrastructure understandable as the system grows.

Agents use explicit tools

Models can reason, but production services still own policy, validation, authentication and data access.

Operability is part of design

Clear failure paths, observable workflows and testable components matter before a feature reaches production.

From automation to production AI.

Each role moved closer to owning the services behind intelligent products.

Full history on LinkedIn, opens in a new tab

Plexus Tech

Conversational banking integration

Software Engineer, BBVA project, Mexico City

2026 - Present

FIXAT

Serverless financial RAG

Software Engineer, Remote

2024 - 2025

Hitss México

API quality for Claro Drive

Backend / QA Intern, Remote

2023 - 2024

BASF

Process automation and analysis

Python Automation / Digitalization Intern, Mexico

2022 - 2023

B.Eng. in Systems Engineering, Instituto Politécnico Nacional.

Let's build dependable AI.

Open to backend and AI engineering roles with U.S. teams.

Email me