UNTAN INFORMATICS · BACKEND & AI DISCIPLINE · @ETHERIFT
Andrean Stephanus
▸▍
Building systems that scale — from high-performance backends to AI with Retrieval-Augmented Generation.
GoPythonC++RabbitMQRAG SystemsMLflow
UNTAN · INFORMATICS
etherift.website
etherift@core:~$ neofetch --brief
role: Backend & AI Engineer
core: Go · Python · C++ · RabbitMQ
flagship: Luko AI (RAG) & Chronicles-OJ
infra: Proxmox · Docker · PostgreSQL · Valkey
● all services operational
GoPythonC++RabbitMQPostgreSQLRAG / LLMMLflowDockerLinuxValkeyKafkaProxmoxRedisscikit-learnFastAPINext.jsGoPythonC++RabbitMQPostgreSQLRAG / LLMMLflowDockerLinuxValkeyKafkaProxmoxRedisscikit-learnFastAPINext.js
01
About Me
✦UNTAN · Informatics Engineering
Engineering scalable backends & grounded AI systems.
Informatics Engineering student at Tanjungpura University (UNTAN), focused on backend engineering and AI engineering.
Building APIs & backend systems with Go, PostgreSQL, and RabbitMQ — plus RAG (Retrieval-Augmented Generation) systems for AI chatbots. I believe great systems are scalable and measurable.
Focus:High-Throughput APIsRAG Knowledge EnginesRabbitMQ Broker
0+
Projects
0
Public Repos
0
Followers
0
GitHub Since
01
Distributed Backend
Go · PostgreSQL · Valkey
Merancang API berkinerja tinggi, database concurrency, connection pooling, dan arsitektur microservices terukur.
Explore stack→
02
AI & RAG Systems
Python · Vector DB · LLM
Membangun pipeline RAG (Retrieval-Augmented Generation) untuk chatbot cerdas dengan grounding dokumen presisi tinggi.
Explore stack→
03
Asynchronous Queues & Infra
RabbitMQ · Docker · Proxmox
Mendistribusikan beban komputasi berat (evaluasi kode & inferensi AI) melalui message broker tanpa membebani web thread.
Explore stack→
02
Skills
Teknologi dan tools yang saya gunakan untuk merancang sistem backend terdistribusi dan pipeline AI.
⚡ Flagship SpecializationCore Daily Tech
Core Language
Golang
Backend & Microservices
Membangun REST API berkecepatan tinggi, goroutine concurrency, connection pool tuning, dan worker backend.
Production Ready
AI Stack
Python
AI, RAG & MLOps
Integrasi model LLM, chunking dan retrieval dokumen (RAG), vector similarity search, dan pipeline CI/CD MLflow.
Production Ready
Distributed Queue
RabbitMQ
Message Broker & Decoupling
Asynchronous task queue, event routing, AMQP protocol, dan pemisahan beban berat agar API tetap ultra responsif.
Production Ready
Low-Level Systems
C++
Systems & Algorithms
Pemahaman mendalam tentang memory management, struktur data kompleks, dan logika sistem kompetitif/penilaian kode.
Production Ready
▸ Full Tech Ecosystem
⚡
Flagship Core Stack
4 toolsGo
AdvancedPython
AdvancedC++
ProficientRabbitMQ
Advanced🗄️
Backend & Databases
8 toolsPostgreSQL
Database & IndexingValkey
In-Memory Cache & Key-ValueRedis
Caching & Pub/SubKafka
Event StreamingMQTT
IoT Messaging ProtocolJava
Enterprise BackendSpring Boot
FrameworkRESTful APIs
API Design & Docs🧠
AI, RAG & MLOps
6 toolsRAG / LLM Integration
Document Retrieval & ContextVector Databases
Embeddings & Semantic SearchMLflow
Model Registry & Trackingscikit-learn
ML Classification & Regr.Jupyter
Exploratory Data AnalysisPrompt Engineering
Grounding & Few-Shot☁️
DevOps & Infrastructure
6 toolsDocker
ContainerizationLinux / POSIX
System Administration & CLIProxmox VE
IaaS Virtualization & KVMGitHub Actions
CI/CD Automation PipelinesGit
Version Control & BranchingJWT & OAuth2
Auth Architecture02.5
Core Engineering Disciplines
Deep technical competencies across distributed backend architecture, AI grounding engines, asynchronous message queues, and cloud infrastructure.
01DISCIPLINE
Distributed Backend Architecture
High-Throughput Concurrency & Systems
Merancang dan membangun arsitektur backend berkecepatan tinggi dengan Go, optimalisasi koneksi database PostgreSQL, dan REST API terukur tanpa kompromi.
Key Deliverables:
- ▹High-Concurrency Go Microservices
- ▹Database Connection Pooling & Tuning
- ▹Low-Latency RESTful & gRPC APIs
- ▹Zero-Downtime Migration Strategies
Go (Golang)PostgreSQLValkeyRedisJWT / OAuth2
02DISCIPLINE
RAG & AI Agent Engineering
Domain-Grounded LLM Knowledge Systems
Mengintegrasikan model kecerdasan buatan dengan sistem Retrieval-Augmented Generation (RAG) berbasis vector database untuk menghasilkan jawaban presisi dari dokumen terverifikasi.
Key Deliverables:
- ▹Document Chunking & Vector Search
- ▹LLM Context Grounding & Prompt Strategy
- ▹Hallucination Mitigation Pipelines
- ▹Domain Chatbot Knowledge Engines
PythonVector DatabasesLLM APIsSemantic SearchFastAPI
03DISCIPLINE
Asynchronous Queue & Messaging
Task Decoupling & Event-Driven Pipelines
Memisahkan komputasi beban berat (seperti penilaian kode otomatis dan inferensi AI) ke worker pool terisolasi melalui message broker RabbitMQ dan Kafka.
Key Deliverables:
- ▹AMQP Direct / Topic Exchange Routing
- ▹Decoupled Background Worker Pools
- ▹Fault-Tolerant Dead Letter Queues
- ▹Real-Time WebSocket State Streaming
RabbitMQKafkaMQTTDocker SandboxingWorker Pools
04DISCIPLINE
Cloud Virtualization & MLOps
Automated Infrastructure & CI/CD Pipelines
Mengelola server virtualisasi mandiri Proxmox VE serta membangun pipeline CI/CD MLflow untuk pelatihan ulang model dan validasi metrik otomatis.
Key Deliverables:
- ▹Proxmox VE Virtual Machine Orchestration
- ▹MLflow Model Registry & Experiment Tracking
- ▹Automated CI/CD GitHub Actions Pipelines
- ▹Linux POSIX Environment Hardening
Proxmox VEMLflowGitHub ActionsDockerLinux
03.5
Architecture Sandbox
Simulasi interaktif alur sistem terdistribusi: dari API Gateway Go, RabbitMQ message queue, hingga worker AI RAG, sandbox judge engine, & Proxmox IaaS automation.
01inspect
Client Request
HTTP/2 REST API
POST /v1/query: "Jelaskan struktur kurikulum Informatika UNTAN"
○ Click to inspect
02inspect
Go API Gateway
Golang Concurrency
Token validation, rate limiter & connection pool handling
○ Click to inspect
03inspect
RabbitMQ Broker
AMQP Topic Exchange
Decoupled async dispatch: routing_key='rag.inference.v1'
○ Click to inspect
04inspect
Python RAG Engine
Vector DB + LLM
Top-k chunk similarity retrieval + document grounding context
○ Click to inspect
05inspect
Valkey / Cache
In-Memory Store
Cached response & session memory with <2ms lookup
○ Click to inspect
Live System Telemetry
Queue: RabbitMQBroker / Store: Valkey + AMQPTotal Latency: 0ms
Klik "Run Simulation" atau klik node untuk menginspeksi alur sistem...
⚡ Real-Time Systems TelemetryConcurrency Benchmark
Go + Valkey vs Standard DB Latency Matrix
Simulasi throughput dan P99 latency comparison under 10,000 concurrent client requests.
Node #124.5x FASTER
Go + Valkey (In-Memory)
Sub-millisecond Key-Value Cache
AVG LATENCY
1.1 ms
P99 LATENCY
1.8 ms
Throughput28,500 req/s
Node #2OPTIMIZED DB
PostgreSQL (Indexed)
Relational ACID B-Tree Lookup
AVG LATENCY
24.2 ms
P99 LATENCY
38 ms
Throughput1,950 req/s
Node #3BASELINE
Uncached REST Proxy Gateway
Raw Gateway Dispatch without Cache
AVG LATENCY
148.5 ms
P99 LATENCY
210 ms
Throughput240 req/s
03
Featured Projects
Curated selection — distributed backend, AI & RAG, cloud IaaS, and MLOps pipelines.
01AI & RAG Pipeline
On ProgressLuko AI
✦ Flagship//Backend Engineer & AI Engineer
AI chatbot for students asking about informatics topics. Integrates LLM models with a RAG (Retrieval-Augmented Generation) system for document-grounded answers.
System Architecture
LLM + RAG Document Grounding + Async Inference Queue
GoPythonLLMRAGRabbitMQVector DB
02High-Concurrency Engine
ActiveChronicles-OJ
✦ Flagship//Backend API — team of 9
UNTAN's official Online Judge. Code assessment system with a Go, PostgreSQL, and RabbitMQ backend architecture.
System Architecture
Worker Pool + Isolated Sandbox + AMQP Task Routing
GoPostgreSQLRabbitMQDocker Sandbox
03IaaS & Virtualization
ActiveREGtee Cloud
//Backend — team of 3
IaaS/PaaS platform built with Go, Proxmox, Valkey, OAuth2, JWT, and a local payment gateway.
System Architecture
Proxmox API + JWT Auth + Local Payment Gateway
GoProxmoxPostgreSQLValkeyRabbitMQOAuth2JWT
04MLOps & CI/CD
PublicWorkflow-CI
//MLOps Engineer
MLflow CI/CD pipeline: automated retraining & deployment of a RandomForest model (breast cancer classification) with GitHub Actions. Metrics: accuracy, F1, precision, recall.
System Architecture
Automated Retrain + Model Registry + Metrics Validation
PythonMLflowscikit-learnGitHub Actions
03.8
Journey & Experience
Milestones in distributed backend systems, AI engineering, and academic contributions at UNTAN.
2026 – Present
2026 – Present ·Independent AI Project
Backend & AI EngineerLuko AI · Flagship RAG Chatbot
Architecting a Retrieval-Augmented Generation (RAG) knowledge engine for UNTAN Informatics students. Integrating LLM models, vector database embeddings, high-precision chunk retrieval, and asynchronous worker queues.
GoPythonRAGLLMRabbitMQVector DB
2026 – Present
2026 – Present ·Universitas Tanjungpura (Team of 9)
Lead API Backend EngineerChronicles-OJ (Official UNTAN Online Judge)
Building high-performance REST APIs for automated code judging across thousands of student submissions. Implemented RabbitMQ task queues and PostgreSQL database concurrency.
GoPostgreSQLRabbitMQDockerREST API
2026 – Present
2026 – Present ·Cloud Platform (Team of 3)
Backend Infrastructure DeveloperREGtee Cloud · IaaS & PaaS Platform
Developing a VM provisioning platform powered by Proxmox APIs, Valkey in-memory caching, JWT & OAuth2 authentication, and local payment gateway integration.
GoProxmoxValkeyOAuth2JWTRabbitMQ
2026 – Present
2026 – Present ·Open Source MLOps Project
MLOps EngineerWorkflow-CI · MLOps Pipeline Automation
Building automated MLflow CI/CD pipelines for automated retraining and deployment of a machine learning classification model (RandomForest) with GitHub Actions.
PythonMLflowscikit-learnGitHub ActionsCI/CD
2024 – Present
2024 – Present ·Universitas Tanjungpura (UNTAN)
Undergraduate StudentInformatics Engineering (Teknik Informatika)
Studying software engineering, distributed systems architecture, advanced algorithms & data structures, POSIX operating systems, and artificial intelligence engineering.
UNTANPontianakComputer ScienceAlgorithms
04
Contact Me
Interested in collaborating or have a question? Send a message!
Send a Message
Available for collaborationPontianak (UTC+7)
Let's build something exceptional.
Interested in collaborating or have a question? Send a message!
Direct EmailPrimary Channel
andreanstephanus@gmail.com
Open for Backend Engineering, Distributed Systems, and AI/RAG Engineering roles & collaborations.