Systems · software · explanation
I work on backend systems, integrations, and automation — especially where boundaries get messy and change becomes hard to reason about. I like making those systems easier to understand, safer to change, and easier to pass on.
01 / Selected work
A few engineering problems I've spent real time understanding, changing, and improving.
Integration gateways, environment promotion, and delivery workflows that stay safe to change.
Integration gateways, environment promotion, and delivery workflows that stay safe to change.
Read case study ↗Specification-driven services, cross-cutting validation, and quality practices in a large product R&D environment.
Specification-driven services, cross-cutting validation, and quality practices in a large product R&D environment.
Open work detail ↗General Ledger operations, ETL-style batch flows, and production discipline in a large retail bank.
General Ledger operations, ETL-style batch flows, and production discipline in a large retail bank.
Read case study ↗02 / Featured writing
Case studies, technical notes, and longer-form writing about the decisions behind the work.
Featured writing · case study
How inspectable configuration and promotion steps reduce risk at the integration edge.
Architecture and reasoning
from the case study
Teams hit friction when the gateway layer is treated as a snowflake: each environment drifts, reviews are manual, and nobody can diff “what changed” between promote steps.
- No customer data, internal hostnames, or unpublished architecture diagrams in public prose.
1. **Model variation explicitly** — environment-specific settings live in named layers; diffs are mechanical, not archaeological.
More upfront modeling pays off when change frequency is high; for rarely touched routes, lighter tooling may suffice. This case study documents the pattern where repeatability won.
Featured writing · case study
Bounded tooling so analysts can explore assessment rules without production access or one-off engineer scripts.
Architecture and reasoning
from the case study
Loan assessment logic changes often start with analysts who understand credit policy but cannot safely experiment in production. Ad-hoc scripts and shared spreadsheets do not scale review, and every engineer becomes a bottleneck for “what if we changed this threshold?”
- No customer or production data in the playground; synthetic and bounded inputs only.
1. **Separate policy exploration from execution** — configurable scenarios with explicit inputs and expected outputs analysts can diff.
A playground is not a full staging clone; some integration behaviors stay mocked. That is intentional: the goal is faster policy reasoning, not pretending the entire core bank runs on a laptop.
Featured writing · case study
Modernizing COBOL-era batch processing toward observable, idempotent ETL — without pretending the mainframe never existed.
Architecture and reasoning
from the case study
A General Ledger organization still depended on a large COBOL batch estate. Teams needed faster change cycles, better observability, and pipelines that could survive retries without corrupting accounting state.
- Financial correctness beats migration speed; partial automation is worse than slow automation with proofs.
1. **Inventory the real batch graph** — which jobs are pure transform, which encode business rules, which are safe to parallelize.
Not every job belonged in wave one. Some stayed on legacy paths until test harnesses and sign-off rituals caught up. The win is a **repeatable migration pattern**, not a single big-bang rewrite slide.
03 / Engineering throughlines
Across different teams and problems, I keep finding the same ideas useful.
Make environment and configuration differences visible instead of hiding them in special cases.
Enterprise product engineering · Applied computer vision thesis system · KrakenD infrastructure-as-code toolkit · Repeatable API-gateway delivery across environments · Legacy batch jobs to SQL pipelines
Trace this pattern ↗Turn one-off delivery steps into something people can understand, test, and repeat.
Ledger systems and data pipelines · Enterprise product engineering · Platform and integration engineering context · Containerized backend demos · Infrastructure-as-code templates vault · KrakenD infrastructure-as-code toolkit · Omega Education Centre enablement · Public knowledge hub · AWS Back-End Academy facilitation · Bangkit Academy instruction and capstone · BNI ODP 334 — IT operations tutor · Cybersecurity curriculum subject-matter expert · DBS Coding Camp 2026 · Rakamin · BNI ODP · Technical code review and structured feedback · Configuration playground for domain experts · Repeatable API-gateway delivery across environments · Legacy batch jobs to SQL pipelines
Trace this pattern ↗Make interfaces, responsibilities, and operational boundaries easier to reason about.
Ledger systems and data pipelines · Enterprise product engineering · Platform and integration engineering context · Accessible mobile engineering (volunteer) · AI and IoT STEM competition mentoring · BINUS Bekasi — Git and GitHub for first-time users · BNI ODP 343 — tutor and capstone mentor · Code in Place · Samsung / Skilvul · ADPList · DBS Coding Camp 2025 · DBS Coding Camp 2026 · Design thinking for hackathon cohorts · Dicoding partner programs · Rakamin · BNI ODP · ADPList mentoring · Introductory programming section leadership · Microsoft Future Ready · Configuration playground for domain experts · Repeatable API-gateway delivery across environments · System assessment before commitment
Trace this pattern ↗04 / Projects
Personal portfolio site built with Astro — structured content, case studies, and a careful split between drafts and what ships publicly.
CLI and templates for environment-specific API gateway configuration with container publish automation.
Web-based physical distancing detector — object detection, deep learning pipeline, and thesis-to-publication path.
05 / Writing
Some ideas need a full case study. Others are useful as a short note. Writing is where I make the reasoning explicit.
06 / Contribution
Teaching, mentoring, and other ways I try to make what I've learned useful to someone else.
Coding Camp, Bangkit, AWS Back-End Academy, Asah, and PIDI Digdaya.
Instructor + mentorDevSecOps instruction and final-project mentoring.
Career mentorStructured career mentoring for vocational-school students.
Community contributionProgramming instruction, engineering mentoring, and community support.
