Skip to content

About

Paul Christian Mandap

I'm Paul — a Computer Science graduate from the Technological Institute of the Philippines, Manila, where I finished Cum Laude with a machine learning specialisation.

My internship at Stratpoint was where data engineering stopped being coursework. I built an ELT pipeline that turned three messy source files into a dimensional warehouse an insurance underwriter could actually price risk from — and spent most of my time on the unglamorous half: quarantining bad rows with reason codes instead of dropping them, making the whole warehouse rebuild from a single trigger, and proving the rebuild was byte-identical rather than assuming it.

Before that, at Globe Telecom, I consolidated 31 scattered spreadsheets into two governed KPI masters and found a flaw in the report's blend logic that had been quietly producing wrong numbers.

Outside work I build systems that run on my own hardware. A personal AI agent platform on a consumer GPU with no cloud model anywhere in it, and a desktop voice assistant that speaks with a Philippine-English accent implemented as a phoneme transformation rather than a trained voice. Both taught me more about measurement than about models: most of what I learned came from benchmarks that disagreed with what I expected.

Paul Christian Mandap

01Experience

Jun 2026 — Aug 2026

Metro Manila, PH

Stratpoint Technologies

Data Engineering Intern

  • Architected an end-to-end ELT pipeline in Airflow, PySpark and PostgreSQL, transforming raw sources into a three-layer dimensional star schema of 20 dbt models.
  • Cut a full-warehouse rebuild to a single idempotent Airflow trigger completing in 1m22s, with byte-identical output verified against the previous run.
  • Reconciled 12,067 field values across a 45,433-record catalog against TMDB, growing usable data 105% and exposing a true failure rate of 35.7% where the raw data showed 30.0%.
  • Optimised a core query 27% faster and cut its index 76% smaller through EXPLAIN ANALYZE plan analysis and partial indexing.
  • Enforced quality at both layers: 104 dbt tests that fail the pipeline run on reconciliation drift, and 65 pytest cases across the Python layer.

Jan 2026 — Apr 2026

Manila, PH

Globe Telecom

Territory Service Quality Assurance Intern

  • Consolidated 31 source sheets into 2 governed KPI masters feeding Looker Studio, standardising 20+ mobile and broadband metric definitions and documenting KPI-to-source lineage.
  • Found and fixed a flaw in the report's blend logic, rebuilding it on one consolidated master and the full 2025 dataset to collapse 3 overlapping tables into a single source of truth.

02Leadership

Aug 2025 — Apr 2026

AWS Cloud Club — TIP Manila

Chief Technology Officer

  • Led a serverless web deployment workshop for 50+ students, delivering hands-on AWS training in VPC, security groups, S3, Lambda and EC2.
  • Mentored 30+ officers and members on cloud computing fundamentals across compute, storage and networking.

03Core technologies

Data Engineering

PythonSQLPostgreSQLApache AirflowdbtPySparkPandasETL / ELTDimensional modelingStar schemaData quality testingDocker

AI & Machine Learning

scikit-learnKerasYOLOv8OpenCVMediaPipespaCyOllamaWhisperLLM evaluationAgent architectures

Cloud & Analytics

AWS (S3, EC2, Lambda, VPC, IAM)Cloudflare PagesPower BI (DAX)Looker Studio

Software

FastAPIWebSocketsREST APIsNext.jsTypeScriptPyQt6Git / GitHubpytest

04Education

Technological Institute of the Philippines

BS Computer Science — Machine Learning

Cum Laude · GWA 1.52

Manila, PH · June 2026

05Certifications

  • AWS Certified Cloud Practitioner

    Amazon Web Services · May 2025 · Valid through May 2028

  • EF SET English Certificate — 78/100

    EF SET · Feb 2026 · C2 Proficient

  • Data Analytics Essentials

    Cisco · Apr 2025

  • NAISC Certificate of Participation

    AI Singapore · May 2025 · Credential ID 1183