Hyoungsu Kim

Backend Developer (Python/FastAPI) | Software Engineer

aaa57403@gmail.com

About

A reliable Backend Engineer dedicated to solving complex operational challenges through scalable architecture and automation. I specialize in building robust systems using Python (FastAPI) and orchestrating intricate business workflows with Temporal.io (Event-Driven Architecture).

My core strength lies in bridging the gap between technical constraints and business needs. I excel at translating ambiguous requirements into efficient, maintainable solutions through proactive communication with cross-functional teams, including non-technical stakeholders. I am highly adaptable and eager to expand my expertise into full-stack development and AI-driven technologies to deliver comprehensive, end-to-end solutions.

Work Experience

Qualitas Semiconductor

Seongnam, South Korea

Software Engineer

Nov. 2024 - now

  • Focused on developing reliable backend services and automating infrastructure for semiconductor design workflows.
  • Strived to improve system stability and developer productivity by identifying potential bottlenecks and applying appropriate architectural patterns.
  • Architecture Evolution: Improved system scalability by transitioning from an initial polling-based design to a Temporal-based Event-Driven Architecture, resolving performance bottlenecks in complex workflow environments.

Pampam

Seoul, South Korea

Backend Developer

Jan. 2024 - Oct. 2024

  • Engineered the backend infrastructure for a time-sensitive auction service, focusing on ensuring data consistency and low-latency communication under concurrent traffic loads.
  • Real-Time Architecture: Replaced HTTP polling with WebSockets and Redis Pub/Sub to achieve bidirectional data streaming. This design enabled instant price synchronization across horizontally scaled servers, ensuring a fair bidding experience for all connected users.
  • Concurrency Management: Designed and implemented mechanisms to handle concurrent access and maintain system integrity during high-traffic auction events.

Education

Kyonggi University

Computer Engineering

Jan. 2018 - Dec. 2024

Skills

Programming Languages

  • Python (Expert)
  • SQL
  • JavaScript (Basic)

Backend & Architecture

  • FastAPI
  • SQLAlchemy
  • Temporal.io
  • Event-Driven Architecture (EDA)
  • WebSocket
  • RESTful API Design

Infrastructure & DevOps

  • Docker
  • AWS (EC2, ALB)
  • CI/CD (GitHub Actions/Gitea)
  • Ansible
  • Nginx
  • Linux

Data & Observability

  • PostgreSQL
  • Redis
  • Zabbix
  • Grafana

Security

  • Keycloak (SSO/OIDC)
  • JWT-based RBAC
  • API Security

Projects

Qualitas PMS(Project Management System)

Nov. 2025 - now

  • Developed the backend for a task orchestration system designed to manage intricate dependencies in semiconductor design processes.
  • Goal: Build a scalable system capable of handling a growing volume of concurrent tasks without performance degradation.
  • Approach: Chose Temporal.io (Event-Driven Architecture) early in the design phase to move away from synchronous polling methods that were expected to cause system bottlenecks.
  • Implementation:
    • Designed workflows using One-off patterns to ensure high availability and minimal resource usage for each task.
    • Decoupled event reception from logic execution using a Producer-Consumer model, ensuring the system remains responsive during traffic bursts.
    • Used a Reverse Indexing strategy to optimize the mapping between external webhooks and internal task states.

Intelligent Monitoring & Alerting System

Aug. 2025 - Nov. 2025

  • Engineered an automated operational platform to standardize infrastructure management and ensure reliable incident response within a restricted air-gapped environment.

  • Problem: Manual server management caused configuration drift, while the existing monitoring system suffered from delivery reliability issues during high-load incidents.

  • Solution: Built a self-hosted automation suite combining IaC (Ansible) for provisioning and Temporal.io for resilient event processing.

  • Implementation:

    • Resilient Alerting: Utilized Temporal Task Queues to isolate alert delivery logic and implemented a throttling mechanism to handle API rate limits (e.g.,Teams), ensuring 100% reliable delivery during traffic bursts.

    • Dynamic Cooldowns: Integrated Redis to manage dynamic cooldown states, effectively suppressing redundant notification for ongoing incidents and reducing operational fatigue.

    • Infrastructure as Code (IaC): Developed Ansible Playbooks to standardize OS configurations and automate the rollout of Zabbix agents across 30+ servers, ensuring consistent environments.

    • Centralized Visibility: Architected a monitoring bridge using Zabbix Proxy to aggregate data from segmented networks into a unified Grafana dashboard.

Qualitas Sheet(Internal Platform)

Nov. 2024 - Jul. 2025

  • Developed a centralized platform to manage internal project documentation and approval workflows.
  • Problem: Lack of a unified system for authentication and access control made it difficult to manage sensitive project data.
  • Solution: Built a secure platform with standardized identity management and a flexible data model.
  • Implementation:
    • Architected an SSO environment with Keycloak to provide a single entry point for internal services.
    • Developed a JWT-based RBAC system within the FastAPI middleware to manage permissions based on department and user rank.
    • Designed a PostgreSQL schema to handle the complex, multi-level hierarchical structures required for corporate approval processes.

Real-time Auction Platform with WebSocket

Jan. 2024 - Oct. 2024

  • Engineered the backend for a real-time bidding service, focusing on low-latency communication and data consistency.
  • Problem: Traditional HTTP communication was too slow for the real-time requirements of an auction closing.
  • Solution: Implemented a bidirectional communication model with efficient message queuing.
  • Implementation:
    • Used WebSockets and Redis Pub/Sub to ensure that bidding events and price updates were synchronized across all users without delay.
    • Designed a message ingestion layer using Redis Lists to buffer bids, ensuring FIFO processing and protecting the database from traffic spikes.
    • Managed the production environment on AWS, utilizing Auto Scaling and ALB to maintain availability during peak traffic periods.

Certifications

AWS Certified Solutions Architect – Professional

Amazon Web Services (AWS)

Aug. 2025

AWS Certified Solutions Architect – Associate

Amazon Web Services (AWS)

May. 2025