Production · 2026Live
Satellyte
Turns public buying signals into enriched, scored leads for B2B revenue teams.
- Next.js
- Fastify
- BullMQ
- PostgreSQL
- Redis
- Playwright
- Docker
- GitHub Actions
Marco Abel Zefanya
Full-stack engineer building production-grade SaaS and AI products, from the interface to the infrastructure it runs on.
Experience
4+
years across engineering and QA
Shipped
5
products, from SaaS to industrial systems
Based in
Jakarta
Indonesia (GMT+7)
Daily stack
- Next.js
- TypeScript
- Node.js
- Fastify
- Python
- PostgreSQL
- Redis
- Docker
- GitHub Actions
(01)
Selected apps
Products I've built and run in production, across B2B SaaS, computer vision, hiring, operations, and industrial monitoring.
AutoGrade (Production)
2025–26Private production system
Grades palm fruit from three live industrial camera feeds and surfaces results to operators in real time.
Private production system
Turns every weighed and graded truck into a priced ticket, a stock receipt, and a daily yield report.
Helps distance-learning university students practise timed exams, review every answer, and drill their mistakes.
Private production system
Follows every box of parts from a printed lot card through each station scan to a live monitoring board.
Private production system
Takes visitors from the next event to its tickets and lineup, and sends every brand brief to a CMS the agency runs itself.
Guides event attendees through the booths on their phone: a short quiz, a PIN from the booth rep, and an event kit once every booth is done.
The live microsite of Pawlo, a pet supplement and vitamin brand: products by category, inquiries from pet owners and clinics, and a CMS editors run without a developer.
Private production system
Helps pediatricians check a child's growth against WHO standards and work out a formula feeding plan in a few steps.
Private production system
Combines applicant screening, voice interviews, recruiter workflows, and usage-based billing in one product.
Private production system
Connects Sales, Purchasing, Inventory, and Production around one consistent operational flow.
Private production system
Moves plant sensor readings from MQTT ingestion to dashboards and scheduled mobile reports.
Private systems are shown with invented demo data only, never client names or company data.Browse earlier projects
(02)
About
I'm a Full-Stack Engineer with 4+ years of combined experience across production software development and embedded software quality assurance.
My work spans B2B SaaS, AI-powered products, computer vision, enterprise software, industrial monitoring, browser automation, billing systems, and production infrastructure. I enjoy taking products from operational requirements through system design, implementation, deployment, and production reliability.
My QA background shaped the systematic approach I bring to debugging, testing, traceability, and resilient software delivery.
Education
Hacktiv8
Full Stack JavaScript Immersive Program · March 2024 – July 2024 · Final Score: 94.2/100
Education
Universitas Gunadarma
Informatics Engineering · 2016 – 2020 · GPA: 3.21/4.00
Languages
Indonesian · English
Native · Working Proficiency
(03)
Services
For teams hiring and for clients with a project. Same engineer, same standards.
SaaS & web apps
From database schema to a deployed Next.js product, built and owned end to end.
AI features & automation
LLM workflows, data enrichment, browser automation, and computer vision that hold up in production.
Dashboards & internal tools
Operator dashboards, ERP modules, and admin panels that turn messy operations into clear flows.
APIs, pipelines & infra
REST APIs, queues, data pipelines, Docker, and CI/CD, so releases stay boring.
Process
- Step 01
Understand
Map the operational problem, the users, and the constraints before writing code.
- Step 02
Design
Shape the data model, APIs, and system boundaries so the product can grow.
- Step 03
Build & test
Ship in small, tested increments. My QA background keeps the edge cases honest.
- Step 04
Ship & monitor
Deploy with Docker and CI/CD, then watch production and keep it reliable.
(04)
Experience
From QA to full-stack engineer.
Highlights
- Develop Next.js dashboards and Fastify/BullMQ services that process multiple buying-signal strategies.
- Build multi-provider enrichment workflows for identifying and scoring decision-makers.
- Implement browser automation with Playwright and Browserbase.
- Develop public REST APIs with authentication, credit metering, and rate limiting.
- Support production delivery with Docker, GitHub Actions, and GHCR.
- Next.js
- React
- TypeScript
- Fastify
- Node.js
- BullMQ
- PostgreSQL
- Redis
- Python
- FastAPI
- Playwright
- Browserbase
- LLM APIs
- Stripe
- Docker
- GitHub Actions
- GHCR
Highlights
- Developed a Python computer-vision service with FastAPI, YOLO, OpenCV, and TensorRT for three industrial camera feeds.
- Built TypeScript APIs, role-based access control, event delivery, and a Next.js operator dashboard for three production lines.
- Designed resilient delivery with idempotency keys, local buffering, and batch synchronization.
- Developed a multi-tenant AI recruitment platform with AI CV screening, voice interviews, candidate pipelines, and Stripe billing.
- Built an MQTT, Node-RED, TimescaleDB, Grafana, and Telegram monitoring pipeline.
- Next.js
- React
- TypeScript
- Express.js
- Python
- FastAPI
- YOLO
- OpenCV
- TensorRT
- PostgreSQL
- MongoDB
- Redis
- Stripe
- MQTT
- Node-RED
- TimescaleDB
- Grafana
- Docker
- GitHub Actions
Highlights
- Built Next.js and React interfaces supported by REST APIs developed with Node.js and Express.js.
- Designed operational data models using PostgreSQL and MongoDB.
- Implemented authentication, role-based access, search, filtering, pagination, exports, and cross-document workflows.
- Translated internal operational workflows into practical product features.
- TypeScript
- Next.js
- React
- Node.js
- Express.js
- PostgreSQL
- MongoDB
- REST APIs
Highlights
- Performed white-box testing on embedded transmission-control software written in C.
- Achieved the required 100% test coverage across assigned software modules.
- Developed unit tests with winAMS and debugged output discrepancies.
- Built Excel-based verification tools and produced traceable test cases, coverage reports, and technical documentation.
- C
- winAMS
- White-box Testing
- Unit Testing
- Microsoft Excel
(05)
Stack
Tools I use in production, grouped by layer.
Programming Languages
- TypeScript
- JavaScript
- Python
- SQL
- C
Frontend
- React
- Next.js
- React Native
- Tailwind CSS
Backend & APIs
- Node.js
- Fastify
- Express.js
- FastAPI
- REST APIs
- GraphQL
- BullMQ
Databases
- PostgreSQL
- MongoDB
- Redis
- TimescaleDB
AI & Automation
- Playwright
- YOLO
- OpenCV
- TensorRT
Infrastructure
- Docker
- GitHub Actions
- GHCR
- DigitalOcean
- Cloudflare
Industrial & IoT
- MQTT
- Node-RED
- Grafana
- Raspberry Pi
(06)
Lab
Public code anyone can read, starting with this site.
web-portfolio (opens in a new tab)
This site. Next.js 14, English and Indonesian copy, an AI chat with rate limiting, and a content editor that commits through the GitHub API.
- Next.js
- TypeScript
- Vitest
(07) Contact
Let's work
together.
Open to full-time roles and freelance projects.
HiringNeed an engineer on your team?
Full-time roles in full-stack, backend, or AI product engineering.
Open to full-time roles and freelance projects.
FreelanceHave a project in mind?
SaaS, AI features, dashboards, or internal tools. Tell me what you need.