Sheet 01. Overview

I build AI agents and automations that ship to production.

Full-stack AI engineer turning ideas into working products: LLM agents, automation pipelines, and Next.js apps users rely on.

View Resume
  1. A

    Input

    Your data and APIs

    documents · databases · webhooks

  2. B

    Build

    LLM agent + pipeline

    RAG · tool calls · evaluation

  3. C

    Ship

    Production app

    Next.js · FastAPI · Supabase

Main view. How every engagement is structured, from your inputs to a deployed product.

Specification

AI features shipped
15+
Less manual processing
80%
Users supported
1,000+
Uptime
99.9%

The problem

A process eating hours of manual work every week, or an AI feature your competitors already shipped.

Hiring a full team is slow and expensive. A demo that only works on stage does not help either. You need someone who can scope it, build it, and keep it running.

  • One engineer, whole stack

    Frontend, backend, AI layer, database, and deployment from one person. Fewer handoffs, faster iteration.

  • Measured results

    My automation pipelines cut manual processing by up to 80% and sped up decision-making by 40%.

  • Built to stay up

    Full-stack systems serving 1,000+ users at 99.9% uptime, designed for high-concurrency automation tasks.

Services

How I can help

  • AI agents & chatbots

    One LLM-powered feature integrated into your product: agents, chatbots, RAG search, or document extraction.

    Python · LangChain · OpenAI and Claude APIs

  • Automation pipelines

    Replace a manual process with web scraping, data pipelines, and API integrations. The same kind of B2B pipelines that cut manual processing by up to 80%.

    Python · FastAPI · Docker

  • Full-stack applications

    Production web and mobile apps with AI embedded where it adds real value.

    Next.js · React · Supabase · Flutter

  • API & backend services

    REST and streaming APIs, database design, and Supabase integration.

    FastAPI · Node.js · Postgres

Case studies

Shipped work, with the result it produced

Each detail shows the problem it solved and what changed once it was live.

ApplAI screenshot
Detail A. ApplAI

75%

less job-search time

ApplAI

View project

Problem

Applicants scroll through listings that do not match their skills, while recruiters screen candidates by hand.

Result

  • 75% less job-search time for applicants
  • 90% less manual searching, matched automatically to skills
  • Recruiters fill roles 50% faster with AI-scored candidate lists
  • 3x fit accuracy for both applicants and recruiters

+320%

engagement actions per campaign in 7 days

Plex

View project

Problem

Growing engagement on a social media account meant relying on organic reach or paying for ads.

Result

  • +320% engagement actions per campaign in the first 7 days versus an organic baseline
  • 60% lower cost per engagement than paid ads
  • 78% task completion rate and 45% 30-day user retention
  • Campaigns launch in under 5 minutes and payouts process in under 24 hours

Built with

Flutter, Dart, FastAPI, AWS, Docker

91%

answer accuracy with the source cited

RAG Marketing Intelligence Platform

Problem

Strategists spent their time on manual research and processing across the agency's campaign data.

Result

  • 40% faster decision-making
  • Up to 80% less manual research and processing time
  • 91% answer accuracy with the correct source cited
  • Median response time under 6 seconds at 99% uptime
  • 85% of strategists use it weekly

Built with

Python, FastAPI, TypeScript, Next.js, LLM APIs, Vector search, AWS, Docker

AWS Mock Exams screenshot
Detail D. AWS Mock Exams

Live

on Google Play

AWS Mock Exams

View project

Problem

Learners preparing for AWS certifications needed practice exams they could take on a phone.

Result

  • Developed and deployed to the Google Play Store
  • Built during my IT internship at Tutorials Dojo

More work

Experience

Where the numbers come from

  1. Jan 2026 to present

    InventivFull-stack Developer

    • 15+ AI-driven features across web and mobile, improving decision-making speed by 40%
    • End-to-end automation pipelines for B2B systems, cutting manual processing time by up to 80%
    • High-availability full-stack systems supporting 1,000+ users at 99.9% uptime

  2. Oct 2025 to Jan 2026

    Tutorials DojoIT Intern

    • Developed and deployed the AWS Mock Exams mobile app to the Google Play Store
    • Authored monthly technical articles and hands-on GCP lab guides

  3. Class of 2026

    University of Southeastern PhilippinesBS Computer Science, Data Science major, Dean's List

    • Coursework in data science, big data, machine learning, and computer networks

  4. Aug 2024 to May 2025

    Computer Science SocietyExternal Vice President

    • Managed partner-organization communications and industry events for members

Technical skills

Languages
PythonTypeScriptDart
AI
LangChainOpenAI APIClaude APITensorFlow
Frameworks
Next.jsReactFastAPIFlutterLaravelTailwind
Infrastructure
SupabaseDockerVercelGitGitHubPostman

FAQ

Common questions

Can one engineer handle the whole project?

Yes. I build features end to end: database, backend, AI integration, frontend, and deployment. You get one point of contact for the whole build.

Will the AI work in production or only in a demo?

The systems I maintain run at 99.9% uptime for 1,000+ users. I deploy on infrastructure you own and hand over the code and documentation.

Are you available for contract work or full-time roles?

Both. I take freelance projects through Upwork or direct contract, and I am open to full-time roles.

How much does a project cost?

Projects are quoted individually. Send a short description of what you need and I will come back with a scope and a price.

Contact

Have a process to automate or an AI feature to ship?

Send a short description of what you need and I will come back with a scope and a price.

Opens your email app with the message filled in.

Or reach me directly