Skip to content
View lucasngucii's full-sized avatar
🏠
Working from home
🏠
Working from home

Highlights

  • Pro

Organizations

@linkadz-org

Block or report lucasngucii

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
lucasngucii/README.md

Le Hoang An  ·  Lucas Aleh

SOFTWARE ENGINEERING   EVENT-DRIVEN SYSTEMS   SYSTEM DESIGN

Fullstack product builder. Mostly backend, mostly distributed systems.

Email Résumé Repositories



About

Four years building products end to end — backend architecture, real-time pipelines, and the UI on top of them.

  • Limzy — backend design for a multi-channel adtech platform handling campaign data in near real-time.
  • Uniscore — real-time data pipeline for a product serving around 1M users.
  • Comfortable with the messy parts: traffic spikes, unreliable upstream data, third-party API failures.

I like joining products early and staying accountable for how they behave in production.


Open Source

A few things I work on outside of client work.

Project What it does Stack
Argus
v0.1 · alpha
A local-first gate for AI coding agents. Parses each command into an AST, scores it by severity, and returns allow / ask / deny. Decisions are stored locally so they can be explained and replayed. Go SQLite
product-hub
contributor
Product operating system for a team — roadmaps scored by RICE, OKRs that roll up on their own, self-generating cycles, and burn-up charts reconstructed from issue timestamps. Multi-tenant, and the workspace itself is an MCP server. NestJS MongoDB React TypeScript
face-detection Attention monitoring from a webcam — MTCNN for detection, MediaPipe landmarks, six geometric features, then a classifier. Around 92% accuracy at ~32 FPS. Python scikit-learn
clickhouse-owl
contributor
Web admin console for ClickHouse — tabbed SQL editor, natural-language query generation, visual table builder. Next.js TypeScript

End to End

The layers I've owned on production systems, and what I reach for at each.

Layer What that means in practice Tools
Product Scoping, tradeoffs, deciding what not to build yet
Interface App architecture, server components, state, design system Next.js TypeScript
API Service boundaries, contracts, auth, versioning against third-party APIs NestJS Node.js
Domain DDD modeling, CQRS, transactional consistency, migrations PostgreSQL Redis MongoDB
Events Async workflows, retries, idempotency, outbox and saga patterns Kafka RabbitMQ BullMQ
Analytics Ingestion, read models, search, ML on top of the pipeline Go ClickHouse Elasticsearch Python
Operations Deploys, observability, on-call, incident response AWS Docker Swarm
Read models are rebuilt from the event log, so a bad projection is a replay rather than an outage. Services get extracted only where the scale profile actually differs — the rest stays a modular monolith.

GitHub statistics Most used languages

lean09062@gmail.com

Pinned Loading

  1. face-detection face-detection Public

    Jupyter Notebook

  2. nestjs-bustransit nestjs-bustransit Public

    Forked from diepnghitinh/nestjs-bustransit

    TypeScript

  3. diepnghitinh/clickhouse-owl diepnghitinh/clickhouse-owl Public

    A monitoring & mangement Clickhouse

    TypeScript 2 1

  4. Argus Argus Public

    Your AI coding agent has a shell — and tools

    Go 2

  5. nodeterm nodeterm Public

    Forked from eneskirca/nodeterm

    TypeScript