Harvy Monte de Ramos
Resume

Software engineer · Full-stack · AI & automation

I build software that solves real business problems.

Full-stack applications, backend systems, and AI-powered automation — built to streamline workflows, reduce repetitive work, and turn ideas into working products.

Full-stack products · APIs & backend · AI workflows · Automation

View selected work Contact me View resume
GitHubLinkedIn Email

Selected work

Evidence of how I build.

I prioritize projects that expose a real system: inputs, decisions, automation, and the interface where the outcome becomes useful.

Want a project demo? Email me

01 Sales intelligence / outbound workflow

Prospectos

An AI-powered prospecting platform that discovers local businesses, audits their websites for verified opportunities, and manages controlled Gmail outreach.

Discovery / qualified local leads

Find a market before writing the first email.

Prospectos searches local businesses by category and location, then makes the discovery set usable with contactability signals, website presence, filters, and Prospect Fit scoring.

Source
OpenStreetMap and Geoapify discovery for category and city searches
Triage
Email-ready leads, website signals, review context, and audit potential
Prospectos local business discovery map and prospect results
Discovery / location, category, and lead filters

A real search area and lead set before a business enters the outreach workflow.

Audit to delivery

Evidence carries through to approval.

A crawl identifies a credible conversion or automation gap, ranks it by confidence, and gives the outbound draft one supported sales angle rather than a generic pitch.

Crawl → verified finding → selected angle → reviewed draft

Prospectos website audit with scored evidence and responsive page captures
Audit / opportunity scoring and evidence

Confidence, supporting signals, and desktop/mobile captures keep the recommendation defensible.

Prospectos Gmail-connected evidence-based outreach draft awaiting approval
Outreach / Gmail-connected review

The audit finding becomes a draft that is reviewed and approved before delivery.

02 B2B SaaS / RAG platform

Day One

A team knowledge platform designed to preserve context when engineers leave and new people need answers quickly.

Primary surface / Ask Repository

Answers grounded in the codebase.

Repository content is ingested, chunked, indexed, and retrieved before the system responds. The interface keeps the source-oriented question flow visible instead of hiding it behind a generic chat screen.

Input
Repository context, architecture decisions, and engineering questions
Output
Grounded answers designed for onboarding and ongoing system understanding
Day One workspace showing source-grounded knowledge retrieval
Workspace / source-grounded knowledge retrieval

Repository-aware questions remain tied to their working context.

Engineering context

Architecture, then access.

The supporting interfaces make it easier to map a codebase and establish the team workspace before knowledge retrieval takes over.

Repository → index → retrieval → grounded answer

Day One architecture intelligence system map
Architecture intelligence / system mapping

Visual relationships help teams reason about the repository as a system.

Day One team sign-in screen
Entry point / team access

An explicit workspace entry point before shared knowledge becomes available.

03 Multi-tenant product

OneTap

A community management platform for resident operations, issue reporting, facility reservations, and notification flows.

Primary surface / Community map

A resident app with a real operations layer.

Residents can report issues where they happen. Staff can track their status, coordinate follow-through, and manage community activity from the same system.

People
Residents, community staff, and administrators
Core flows
Issue reports, reservations, notices, and account access
OneTap community map and issue reporting operations
Community map / issue reporting

Location and status turn individual reports into an operational view.

Operational coverage

From resident access to the daily workspace.

The product separates the resident experience from the staff-facing workflow while keeping both sides connected to the same community data.

Resident account → report or reserve → staff follow-through → notification

OneTap resident sign-in screen
Resident access / onboarding

A clear entry point for people using the community platform.

OneTap community workspace
Community workspace / daily operations

Shared updates, issue visibility, and coordination for the community.

Aria / product concept

Meet Aria.

A context-aware assistant for the work already in motion

Aria interprets a request, retrieves the context behind it, and returns the next useful response across projects, tasks, updates, and conversations.

Understands
Classifies intent and asks for clarification when confidence is low.
Connects
Routes a request to project status, task management, agenda, or conversation context.
Remembers
Uses session structure and vector-embedded memory chunks in Supabase pgvector.
Request→Intent→Context retrieval→Useful response

In development · The portrait represents the product’s presence; the example describes the intended interaction model.

Incoming requestWhat changed in the Day One architecture?
→
Aria retrievesRecent project context, active tasks, and relevant memory.
→
Response stateA concise update with gaps surfaced before it answers.

Technical skills

Tools grouped by the work they support.

A complete working stack, organized by the responsibility it helps me own.

01

Frontend and product UI

React, Next.js, TypeScript, JavaScript (ES6+), Tailwind CSS, shadcn/ui, HTML5, CSS3, Three.js, Framer Motion, GSAP, D3.js, and React Flow.

02

Backend and AI

Python with FastAPI, PHP with Laravel, LangChain, RAG pipelines, Groq API, LLM integration, prompt engineering, n8n automation, and REST APIs.

03

Databases and cloud

PostgreSQL, MySQL, Supabase (Auth, Realtime, Storage, and pgvector), DynamoDB, AWS Amplify, Vercel, and Railway.

04

Delivery, quality, and tools

Claude, Claude Code, Cursor, GitHub Copilot, Git and GitHub, Agile/Scrum, iOS/Android/React web QA, Microsoft Clarity, Resend API, Figma, and Wix.

Credentials

Education and credentials.

Academic work, recognized commitment, and training across applied AI, cloud operations, and software engineering.

Education2022 - 2026

University of San Jose-Recoletos / Cebu City, Philippines

Bachelor of Science in Information Technology

Web Applications Development

  • Submitted BantayKo, an AI personal safety app, and a barangay community digitization platform to Solutionsfest 2026.
  • Bronze Loyalty Award recipient.
  • Participated in team-based sports and esports competitions throughout university.
01

IBM / Coursera

Generative AI for Growth Marketing

Generative AI applications, prompt engineering basics, and AI-driven growth marketing strategy.

02

Amazon Web Services

AWS Academy Graduate - Cloud Foundations

Cloud fundamentals, core AWS services, and foundational cloud architecture concepts.

03

Amazon Web Services

AWS Academy Graduate - Cloud Operations

Cloud operations, monitoring, reliability, and day-to-day service management.

04

Alliance Software Inc.

Alliance Software Engineering Program

React, JavaScript, TypeScript, .NET, and C# software engineering training.

Experience

Recent work, with clear ownership.

End-to-end product work, engineering delivery, and cross-functional contribution.

Collaborative project

Lead Backend Developer

Woogidex / Collaborative project

Partnered with the project owner and frontend developer to build Woogidex, a browser-based Fakemon creation platform. Led backend development using Supabase, including Edge Functions, database migrations, identifier-based authentication, login security, and data services for community, contest, and notification features.

Supported real-time multiplayer functionality through WebRTC networking and application protocols, while integrating cloud features with local-first IndexedDB storage.

Lead: Supabase Edge Functions / database design / secure auth / community features / WebRTC / IndexedDB

Mar 2026 - Jun 2026

Software Developer Intern

Xeleqt Technology Innovations Inc. / Cebu City

Rebuilt the company website from Wix to Next.js and shipped it on AWS Amplify. Integrated an AI chatbot and Microsoft Clarity, then contributed QA across iOS, Android, and React web products. Also supported digital-twin UI prototypes, 3D asset testing, and Agile delivery.

Shipped: Wix-to-Next.js migration / AI chatbot / AWS Amplify / Clarity / cross-platform QA

Harvy H. Monte de Ramos

About

Harvy Monte de Ramos

I'm a full-stack developer and AI builder who likes the connective work of product development: the interface people use, the services that support it, and the automation that removes needless effort.

I work comfortably across React and Next.js, FastAPI and Laravel, and PostgreSQL or MySQL—choosing the shape of the system around the problem rather than the other way around.

Start a conversation