Reynald Ace Pilpil

AI / ML Engineer · Agentic LLM Systems & Document AI

I build production-ready ML systems — from RAG assistants and agentic LLM workflows to multimodal document AI. Every project ships to Hugging Face Spaces with a live demo you can click.

View Projects Download Resume Get In Touch

10 deployed projects 4 agentic systems End-to-end RAG Document AI

About

I'm an aspiring AI/ML Engineer who ships production systems, not just experiments. My work spans agentic LLM pipelines, RAG architectures, and multimodal document AI — all deployed with live demos on Hugging Face Spaces.

I came up through fullstack web engineering — Django, MySQL, REST APIs at production scale (1000+ users on my thesis project) — and have spent the last year going deep on applied AI/ML, building systems that move beyond notebooks to real-world deployments.

Currently focused on: agentic LLM systems, RAG architecture, and multimodal document AI.

Cavite, Philippines TUP–Cavite · Computer Engineering Open to remote roles

Technical Skills

Machine Learning & AI

Python PyTorch Hugging Face Transformers BERT DistilBERT LayoutLMv3 Flan-T5 scikit-learn XGBoost SHAP sentence-transformers ChromaDB NumPy pandas

LLM & Agentic Systems

LangChain LangGraph RAG Architecture Multi-Agent Orchestration Groq · Llama 3.3 70B Tavily Search API Pydantic v2 Prompt Engineering LangSmith

MLOps & Deployment

FastAPI Docker Hugging Face Spaces Gradio Railway REST API Design Swagger / OpenAPI pdfplumber python-docx

Backend & Data

Django Flask MySQL SQL Query Optimization JavaScript (ES6+)

Tools & Practices

Git / GitHub Agile Code Review Testing Debugging

Projects

Agentic

HireMap PH — Job Market Intelligence

Full-Stack Philippine Labor Market Platform

A Philippine job market intelligence platform with a live choropleth heatmap and Skill Gap Analyzer. GitHub Actions scrapes 3 job boards every day at 6AM PHT, commits fresh data, and triggers a Railway auto-redeploy — one Groq call generates the daily AI market summary, cached for every user at zero per-user API cost.

Python FastAPI React 19 Plotly Groq (Llama 3.1 8B) GitHub Actions Parquet Railway

GitHub Demo Case Study

Agentic

SentinelPH — Dengue Risk Monitor

Multi-Agent Public Health Intelligence Platform

A full-stack public health intelligence platform for Philippine dengue surveillance. Ingests 12 years of case data, weather records, news signals, and Google Trends; runs an XGBoost risk model that outperforms the ARIMA baseline by ~40%; delivers citation-enforced RAG answers from WHO guidelines; and generates AI situation briefings via a 4-node LangGraph validator workflow.

Python LangGraph XGBoost ChromaDB Groq (Llama 3.3 70B) Streamlit BAAI/bge-small-en-v1.5

GitHub Demo Case Study

RAG

CodeLens — Chat with Any GitHub Repo

RAG Tool for Codebase Exploration

Clone any public GitHub repo and ask it questions in plain English. CodeLens indexes the entire codebase into a vector database, retrieves the most relevant code chunks, and generates answers with file-level citations — so you always know exactly where the answer came from. Built because reading a new codebase cold is miserable.

Python LangChain ChromaDB Groq (Llama 3.3 70B) sentence-transformers GitHub API Gradio

Space GitHub Case Study

Agentic

Market Intelligence Agent

Stateful Multi-Agent System with Self-Revision

An autonomous research agent with a self-evaluating revision loop using LangGraph — agent critiques its own draft, scores it, and re-searches targeted gaps until quality threshold is met. Researches a company across 4 parallel angles (overview, news, competitors, market trends) via live web search and delivers a structured intelligence report as a Word document.

Python LangGraph LangChain Groq (Llama 3.3 70B) Tavily Search API Pydantic v2 Gradio LangSmith

Space GitHub

Agentic

Job Application Assistant Agent

Autonomous Multi-Step LLM Agent

A 9-step agent with integrity enforcement — architecturally prohibited from fabricating resume content. A post-tailoring diff check flags violations, with every change logged in a fully auditable changelog (reason + confidence score). Pipeline: resume parsing → JD scoring → company research → ATS-tailored resume + cover letter + outreach in under 30 seconds.

Python Groq (Llama 3.3 70B) Tavily Search API Pydantic v2 Gradio python-docx

Space GitHub

RAG

Philippine Labor Law RAG Assistant

End-to-End Retrieval-Augmented Generation

End-to-end RAG pipeline over the Philippine Labor Code — document ingestion, semantic chunking, vector embeddings (all-MiniLM-L6-v2), ChromaDB retrieval, and Flan-T5 generation. Returns natural-language answers (leave entitlements, overtime rules, termination procedures) with source citations grounded in the actual law, reducing hallucination through retrieval-first design.

Python LangChain ChromaDB Hugging Face Transformers Flan-T5 sentence-transformers Gradio

Space Colab

Document AI

Financial Document Intelligence API

Production REST API

Production-grade REST API that ingests bank statement PDFs and returns structured JSON — extracted transactions, auto-categorized spending, computed financial summaries (income, expenses, savings rate), anomaly flags, and actionable insights. Containerized with Docker, deployed to Railway with auto-generated Swagger / OpenAPI docs.

Python FastAPI pdfplumber pandas Docker Railway

Space GitHub API Docs

Document AI

Receipt Data Extractor (LayoutLMv3)

Multimodal Document AI

Extracts structured data (menu items, quantities, prices, totals) from receipt images using LayoutLMv3, a multimodal transformer that jointly reasons over text, layout, and visual features. Achieved 93.96% F1 on validation set, demonstrating document AI capability beyond plain-text NLP.

Python Hugging Face Transformers LayoutLMv3 PyTorch

Space

Classical ML

BERT Sentiment Analyzer

Social Media Intelligence Tool

Multi-input bulk sentiment analysis tool that processes batches of reviews, tweets, or comments and returns an aggregate sentiment score, breakdown chart, and the top positive / negative phrases driving the score. Dashboard-style output (not single-label classification), making it usable as a social-media intelligence tool.

Python Hugging Face Transformers PyTorch DistilBERT Gradio

Space Colab

Classical ML

Customer Churn Predictor

Customer Retention Dashboard

Production-ready churn risk dashboard: ingests CSV customer data, predicts churn probability with XGBoost, and outputs a retention report with risk distribution, ranked high-risk customers, and SHAP-based feature explanations — translating raw model output into business-actionable retention recommendations using explainable AI (SHAP).

Python scikit-learn XGBoost SHAP pandas Gradio

Space Colab

Experience

Technical Sales Intern

3E HITECH SOLUTIONS, INC.

January 2026 – June 2026

Supported pre-sales activities for the company's solutions, focusing on projects with the National Grid Corporation of the Philippines (NGCP) — contributing to proposal preparation, scoping, and technical clarifications with stakeholders.
Assisted project management across NGCP deployments — coordinating timelines, tracking deliverables, and supporting communication between technical teams and clients.

Certifications