Juan J. Martinez

Lubbock, TX · (915) 777-5465 · [email protected]

I am a Senior Software Developer Team Lead at Quext, a company that provides chatbot solutions for the multifamily industry. I have over five years of experience in the entire product development cycle, from ideation and planning to design, implementation, and maintenance.

My passion is to bring conversational AI to life, using cutting-edge technologies such as vector embeddings, semantic search, GPT LLM, and Dialogflow. I have led and contributed to several successful projects that have enhanced the company's chatbot product, generated positive feedback from clients and users, and created new business opportunities. I have also designed and developed REST APIs, relational databases, and front-end applications using various frameworks and tools.

I hold a Bachelor's degree in Computer Science from The University of Texas at El Paso, where I graduated in 2018. I also won first place in the Parkhub Challenge, a hackathon that required creating a solution to predict future traffic conditions using historical and public event data.

As a team leader, I am adept at bringing fresh ideas to the table and communicating effectively with both technical and non-technical managers. I value collaboration, innovation, and quality, and I strive to deliver solutions that meet and exceed the expectations of the stakeholders and the end-users.

Experience

Lead Software Developer

Madera Residential/Quext
  • Wrote technical aspects of a white paper that allowed the company to be part of the AWS Generative AI Innovation Center program.
  • Architected solution to searching and retrieving documents for Retrieval Augmented Generation (RAG) solution using a combination of a relational and vector database that supports multi-property conversations.
  • Led the design of a relational database to expand the chatbot’s capabilities to multi-property conversations, which increased the number of apartment complexes that use it to more than fifty.
  • Promoted and implemented a solution to enhance the company’s chatbot product by using an AI framework named Retrieval Augmented Generation (RAG) to allow the expansion and usage of domain- specific data. A demo was presented by the company’s CEO at the Texas Apartment Association annual conference, effectively promoting the product and generating positive feedback from the audience.

    Digital Human product on a website

Jun 2020 - Present

Software Developer

Madera Residential/Quext
  • Architected and developed a Dialogflow webhook to support real-time unit availability and tour scheduling pricing conversations that have generated thousands of leads.
  • Contributed to training a Dialogflow agent that responds to more than 350 questions from potential apartment residents, including tour scheduling capabilities and unit availability information.
  • Designed and developed a vue.js single-page chatbot application with a realistic CGI leasing agent and other interactive components that support thousands of monthly conversations.
July 2019 - June 2020

Software Developer

VIMA Transportation Inc
  • Designed and Developed an Android app to provide freight information to truck drivers that also features background geolocation to provide freight updates via email to logistics companies such as C.H. Robinson and JB Hunt, which effectively reduced the number of location-related inquiries by 80%.
  • Developed Smart Fleet Manager, a single webpage application used by fleet dispatchers to dispatch and manage load data and provide load updates to freight brokers.

    View Project

June 2017 - December 2018

Education

The University of Texas at El Paso

B.S. in Computer Science
Fall 2018

Professional Development

Certificate in Architecting in AWS

Amazon Web Services
June 2020

Awards & Special Projects

1st Place Parkhub Challenge

HackDFW

Won first place out of twenty-five teams by developing a solution in less than 24 hours to predict traffic conditions using historical traffic and public event data to allow users to find parking spaces for sporting events quickly. I contributed to the team by developing the mobile app for end users.

February 2019

ARound

HackTX

Developed an augmented reality android application designed to provide customers with 3D models, reviews, and detailed information about products using existing store tags found on shelves in physical stores. This project was featured in a hackathon competition named HackTX, where it achieved a top 10 position out of 80 teams.

Project URL

October 2018

Free Ebay Templates

Personal Project

Won first place out of 25 teams by developing a solution that makes use of past traffic data and public event data to predict future traffic conditions. My contribution to the project was to develop an android application that provides the user a route to a space available near a public event (Kotlin, Java, XML, Google Maps API).

GitHub Repository

February 2016 - June 2016

Skills

Programming Languages & Frameworks
  • TypeScript
  • Python
  • Kotlin
  • PHP
  • Vue.js
  • Nest.js
Other technologies
  • Pinecone (vector database)
  • Open AI API (GPT 3.5/4.0 LLM and ADA embedding model)
  • Google Dialogflow SE (NLP engine)
  • HTML5
  • CSS
  • AWS (cloud computing)
  • Jenkins (CI/CD tool)
  • PostgreSQL
  • MySQL