Joseph Karl Crisostomo
Manila, PH (UTC+8) · joseph.dev.ph@gmail.com · +63 960 4431 908
Portfolio: https://kathulhur.github.io · GitHub: https://github.com/kathulhur
Summary
Python developer and full-stack engineer with hands-on experience building
cloud-native services, AI/ML deployment pipelines, and automation scripting. I
work across Python/Django backends and containerized cloud infrastructure
(Docker, Kubernetes, CI/CD), and I pick up new platforms quickly — eager to
extend my cloud experience into Microsoft and Azure. I work iteratively
(requirements → prototype → deploy → feedback) and collaborate easily with data
scientists, DevOps, and non-technical stakeholders.
Experience
Backend Developer — Department of Science and Technology, Advanced Science and Technology Institute (DOST-ASTI)
On-site · Feb 2024 – March 2026
- Developed and maintained Python/Django backend services for an AI platform,
building automation scripts and the workflows that expose ML models as APIs.
- Built an automated build/deploy pipeline (source in S3, container builds in
CodeBuild, images in ECR) — the kind of CI/CD automation that maps directly to
GitHub Actions / Azure DevOps.
- Managed containerized (Docker) services and orchestrated them with Kubernetes
(AWS EKS), deployed across on-premise Linux (Ubuntu) and the cloud.
- Automated infrastructure provisioning with CloudFormation for fast, repeatable
deployments.
- Optimized database queries and Python workloads through in-depth inspection and
root-cause analysis, improving performance and scalability.
- Integrated cloud data services: managed relational database, Redis caching, and
a message broker for asynchronous workflows.
- Initiated unit testing from 0 to ~90% code coverage; ran automated smoke and
load testing with Grafana K6 to enforce code quality and reliability.
Web Developer (Intern) — Dashlabs.ai
Remote · Aug 2022 – Sept 2022
- Contributed to the maintenance of the platform's Patient Management Service.
- Handled customer queries across multiple channels (email, chat).
Projects
AI-as-a-Service Platform (DOST-ASTI)
Defined the foundational elements of AI development through deployment —
including a model/dataset repository and a feature to easily expose hosted models
as services. Standardized how models are developed, deployed, and served,
cutting AI-model integration from ~2–3 months to a matter of days. Built with
Python; hand-written where it counts.
A community app crowdsourcing near-real-time electricity grid status so citizens
can see outages at a glance. Piloted an agentic-AI build (Claude Code), using
raw-coding experience to catch when the AI drifted off course. Onboarded several
community "scouts" and gained strong local support.
Skills
- Languages: Python, Django, TypeScript, JavaScript, Ruby, Ruby on Rails
- AI/ML & Data: ML model deployment pipelines, AI-as-a-Service platform,
database query optimization, Redis caching, message broker (async workflows)
- Cloud / DevOps: Docker, Kubernetes (EKS), CI/CD (CodeBuild, ECR),
CloudFormation, AWS (S3, RDS, ElastiCache, MQ), on-premise Linux (Ubuntu);
eager to apply to Azure
- APIs / Tooling: REST integrations, Git-based workflows, automation
scripting, Playwright, Grafana K6, WebSocket/SSE
- Strengths: Adaptive (learns new tools fast), technology-agnostic, strong
communicator across varied backgrounds
Education
B.S. in Computer Science — Polytechnic University of the Philippines
2019 – 2023