Senior Backend Engineer | Golang & Python | AI Infrastructure & Distributed Systems
I'm a backend engineer with 5+ years of experience building high-throughput systems and, more recently, production LLM agent infrastructure — mainly in fintech/lending and customer-service automation.
🌐 Portfolio: hairul.web.id
- Design and build scalable backend services in Golang and Python
- Build production LLM agent workflows (LangGraph, RAG, tool-calling) with tracing and latency evaluation
- Design event-driven, idempotent systems on Kafka with strong data consistency guarantees
- Optimize system performance — cut a data-sync job from 2.5 hours to 15 minutes with concurrent worker pools
- Build credit-scoring and decision-engine integrations for loan origination platforms
- Migrate legacy monoliths into maintainable microservices
Backend
- Golang (Echo), Python (FastAPI), TypeScript (Bun runtime)
- PHP (legacy systems & migrations)
AI / LLM
- LangGraph, LangChain, LangSmith (tracing & prompt evaluation)
- RAG: chunking, embeddings, reranking, semantic caching
- MCP servers, n8n automation, Google Vertex AI / Google GenAI
Architecture & Infra
- Microservices, RESTful APIs, hexagonal/clean architecture
- Kafka (event-driven, idempotent processing)
- Docker, CI/CD, GCP
Database
- PostgreSQL (tuning, pgvector), MySQL, SQL Server, SQLite
- Redis, Drizzle ORM
- Scaled a data ingestion pipeline 10x using concurrent Golang worker pools
- Cut core API latency ~40% through SQL tuning, caching, and asynchronous processing with Kafka
- Built credit-scoring and decision-engine integrations for a loan origination system
- Cut LLM agent response latency ~75–80% with vector caching and semantic search (pgvector, Redis)
- Migrated legacy PHP monoliths into Golang microservices
- Production LLM agent systems (LangGraph, RAG, tool-calling)
- Go concurrency patterns for high-throughput data pipelines
- AI-assisted engineering workflows (MCP, multi-LLM orchestration)