AI & Machine Learning
Classification and regression models, owned through evaluation, inference, and retraining.
AI/ML Engineer · Software Engineer
I build practical AI systems end to end: machine learning models, computer vision pipelines, and the data and software infrastructure that carries them into production, plus AI-enabled prototypes on edge hardware.
01About
I'm Mairene Sumalague (Maye to most people), a machine learning and software engineer. I lead ML delivery at Triseed Software Inc, building classification and regression models, GenAI and agentic workflows, and the automated pipelines, retraining jobs, and dashboards that keep them healthy in production.
I started by teaching myself HTML, CSS, and JavaScript after class in junior high, because my school didn't offer programming, and later picked up Java and Python on my own time. Computer Engineering at Batangas State University gave me both ends of the stack: microprocessors and digital logic on one side, software, data, and applied AI on the other. That mix still shapes my work. I'm as comfortable tuning a model as I am wiring a Raspberry Pi or Jetson into a prototype.
I like taking ownership, learning fast, and shipping systems that are reliable, well structured, and easy for a team to maintain.
At Triseed I own models from first look at the data to monitoring in production. Every stage feeds the next, and monitoring feeds back into the start.
02Projects
GenAI platforms, applied machine learning, computer vision, edge hardware, and the software around them. Public repositories are linked. Client work stays private: clients are described generically and shown with original illustrations instead of real data.
Project details could not load. Browse the public repositories on GitHub: ThyroCare app, ThyroCare ML notebooks, and IoT device type identification.
03Skills
Grouped by how I use them. Drawn from production work, freelance projects, the repositories above, and certification work.
Classification and regression models, owned through evaluation, inference, and retraining.
Production GenAI systems: retrieval over private data, agentic workflows, multimodal vision LLMs, and evaluation, built on Claude, OpenAI, and Gemini.
Detection pipelines from dataset checks to real-time inference and operator-facing output.
Python for ML and data work, SQL for most things data, and the rest where a project needs them.
Backend services, secure internal web apps, and the front ends that put models in front of people.
Scheduled ingestion, ETL, and reporting pipelines, plus dashboards that track KPIs and model health.
AI-enabled hardware prototypes, from flight-controller tooling to onboard compute.
Shipping and running systems: containers, CI, observability, testing, and workflow automation.
04Experience
Production machine learning, data systems, and software, plus the education and certifications behind them.
Nov 2023 – Present
Sep 2022 – Oct 2023
Jan 2022 – May 2022
Aug 2018 – Oct 2022
Batangas State University, The National Engineering University
Coursework and projects spanning microprocessor design, HDL-based digital logic, and the foundations of electronics and computer engineering, alongside software systems, data, and applied AI.
Jun 2016 – Apr 2018
STI College Lucena
Science, Technology, Engineering, and Mathematics track with a strong focus on math and science fundamentals.
Anthropic · September 2026
Passed Anthropic's proctored exam for designing and building production-grade applications with Claude, using Claude Code, the Claude Agent SDK, the Claude API, and MCP.
Verify on Credly (opens in a new tab)Oracle · October 2025
Data preparation, model training, evaluation, and deploying ML solutions with Oracle Cloud services and practical data science workflows.
Oracle · October 2025
Building GenAI solutions on OCI, including prompt design, retrieval-augmented workflows, and responsible deployment practices.
Oracle · September 2025
Core AI concepts and OCI fundamentals, including model lifecycle basics, common ML use cases, and best practices for cloud-based AI projects.
05Contact
I'm happy to talk about machine learning engineering, AI-powered applications, data pipelines, and edge AI, whether it's a role, a collaboration, or a hard problem you're working through.