Mike Tran

Full Stack Web Developer | ML Enthusiast

I like building clean, functional, and complex software systems.

Get in Touch

Work Experience

University of Texas at Arlington

Graduate Research Assistant - IDIR Lab

May 2024 — Present

Developing Soil Organic Carbon Knowledge Graph for accurate soil carbon data, funded by NSF and USDA.
Building a real-time dashboard with Streamlit to visualize and analyze soil carbon data.
Building an LLM-based agent system for natural language querying of a knowledge graph.
Maintaining the project infrastructure (database, dashboard, demo website,..)
Co-authored a paper on web engineering and LLMs, accepted at ICWE 2025.

USDA-Agricultural Research Service

Research Intern

May 2025 — Aug 2025

Developed a end-to-end machine learning solution for corn ear phenotyping, including data annotation, model training, and web-based inference.
Created a semi-atomated data annotation pipeline using image processing, reducing annotation time by 90%.
Fine-tuned Faster-RCNN for corn kernel counting, achieving 99% accuracy and 97% recall.
Developed a clustering-based algorithm for organizing corn kernels into rows.

Nagarro

Customer Research Intern

May 2023 — Aug 2023

Analyzed brand sentiment across social media channels using Brandwatch, compiling insights for strategic reports.
Designed and deployed interactive dashboards to present sentiment data and campaign performance.
Created stakeholder-facing reports using storytelling to communicate key insights.

Featured Projects

SOCKG Dashboard

Dashboard for Soil Organic Carbon Knowledge Graph.

Source Preview

Corn Ear Phenotyping

Machine learning solution for corn ear phenotyping.

Source Preview

SPARQL Profiler

Unlike SQL databases, SPARQL doesn’t enforce strict schemas. Having a tool that can infer structure just from an endpoint is surely helpful.

Source Preview

Skills & Expertise

Programming Language: Python, C++

I have around 3 years of experience using Python for a wide range of tasks, from solving LeetCode problems to developing coursework and programming projects.

Web: Streamlit, NextJS, TailWind, FastAPI

I have 1 year of experience as a web developer, with a solid understanding of the full development pipeline—from building backend APIs and frontend UIs to deployment.

Machine Learning

I have built a couple of LLM-backed agentic systems using LangGraph and LangChain. I have internship experience in Computer Vision, specificially in data labeling and fine-tuning Faster R-CNN models. As a GRA in an NLP research lab, I am exposure to NLP and a solid understanding of the field and current development.

Database: RDF, NoSQL, and SQL

I have experience with SQL through coursework and with NoSQL by building a vector database for a RAG system using MongoDB. I also have over 1 year of experience working with knowledge graphs and RDF databases.

Infrastructure & Networking

I have over 1 year of experience setting up and maintaining applications on linux servers, in which I made extensive use of Docker and Docker Compose.

DevOps & Cloud

I gained exposure to DevOps and Cloud through coursework and have successfully migrated several personal projects to AWS, using S3 buckets and ECS clusters with Fargate.

About Me

Hi, I'm Mike Tran (also known as Huy). I'm passionate about web development, AI model architecture, and DevOps practices. I enjoy building clean, robust, and maintainable systems. A well-written codebase, even if complex at first glance, is a work of art to me. My experience spans research, machine learning, web development, and cloud technologies. I'm also a novice photographer and gamer in my free time.