Structural Engineering 

Design, build, and test work on structures and the materials they are made from.

IN PROGRESS: Structural Analysis I (CIVE 3324, Fall 2026). Project work will be added once the course work is complete.

Status key: Completed = finished, with a deliverable · In Progress = current coursework or work underway · Planned = defined, not yet started 

ASCE Timber-Strong Structural Design-Build 

COMPLETED · ASCE UTRGV Chapter · Competed April 2026 · 3rd Place

The challenge: design and build a two-story timber structure to ASCE Timber-Strong competition rules, then load-test it on competition day against the other entered teams.

My contribution

  • Design: helped with the roof design load calculations
  • Detailing: worked through the framing and connection layout with the design team
  • Build: one of the builders translating the drawings into the structure

Process: Design → Calculate → Detail → Build → Load test → Compete

Result: 3rd Place, ASCE Frontier Student Symposium, Corpus Christi, TX (April 2026).

What I learned: turning a load calculation into a buildable connection detail took as much coordination and troubleshooting as the engineering itself. 

Materials Testing Lab

COMPLETED · CIVE 3440 Civil Engineering Materials · Fall 2024 

Hands-on testing of construction materials: how strength, failure, and gradation are measured in the lab.

  • Concrete: batched specimens from aggregate, sand, and cement, cast them in cylinder molds, and loaded a concrete/masonry block to failure in a Gilson compression machine.
  • Wooden truss: a timber truss was loaded to failure on the same machine, with a peak digital readout of 9,780 at failure.
  • Aggregate gradation: sieve analysis on a mechanical shaker to sort aggregate by particle size.

What it shows: three kinds of evidence from one lab: concrete behavior at failure, failure of a built structural member, and characterization of the aggregate that goes into a mix. 

Concrete Crack Detection

RESEARCH · Team project · UTRGV CURE Program 

Problem: structural inspection often depends on a person spotting cracks in concrete. Could an image-classification model flag cracked and non-cracked surfaces automatically?

Approach: a convolutional neural network built with Keras (Python, beginner level) takes a concrete image, extracts features through three convolution and pooling stages, and outputs a single cracked / non-cracked decision.

My role: sourced an existing image-classification model for the team and adapted it (not built from scratch), ran it through 200+ training and testing cycles, and processed image datasets and analyzed results with the team.

Result: about 97% accuracy on a limited test set of roughly 80 images or fewer. This is an early experimental result, not a validated inspection model.