EK

Edward Kim

Available to hire

I’m Edward Kim, a Python engineer who builds the reliable systems and evaluation harnesses that make AI features trustworthy in real products.

About

I’m Edward Kim, a Python engineer who builds the reliable systems and evaluation harnesses that make AI features trustworthy in real products. I focus on observability for model calls end-to-end—capturing inputs, outputs, decision paths, token counts, costs, and outcomes—then using the same trace records for debugging, monitoring, and spend control.

I also ship production integrations and developer tooling: LLM services with REST/SSE/WebSocket endpoints, tool and function calling, and safety guardrails like rate limits, prompt-injection defenses, and approval gates. From voice agents that handle live customer calls around the clock to offline, deterministic CI evaluation suites, I work to ensure performance, reproducibility, and correctness across cloud environments.

  • 22 years' experience
  • Joined September 2026

Industry experience

  • Software & Internet
  • Healthcare
  • Media & Entertainment
  • Financial Services

Work experience· 6 roles

Sole Architect and Developer · CivactaJan 2026 – Present · 11 mos

Designed and developed production systems through discovery-to-production, including CI access control models enforced at the API route level, FastAPI services with async SQLAlchemy/asyncpg/PostgreSQL and Redis integration, JWT access with refresh, and automated audit logging for agent decisions. Implemented content moderation fallback logic with weighted review queues and human thresholds; enforced global spend and per-user quotas before calls.

AI Quality Reviewer and Project Consultant · GreenLight.aiJan 2026 – Present · 11 mos

Contracted as an AI and human-in-the-loop review layer for LinkedIn training initiatives. Scored outputs against a structured rubric and wrote per-score justifications; promoted into the review layer to sign off tasks before final submission.

Senior Software Engineer, Agentic Coding (AI Quality / Evaluation) · AlignerrJan 2026 – Present · 11 mos

Evaluated AI-generated code for correctness, efficiency, and security; built coding problem stress tests for autonomous agents; contributed to evaluation rubrics and frameworks used to judge agent behavior.

Selected Builds (RAG / LLM Products and App Development) · Selected BuildsJan 2024 – Present · 2 yrs 11 mos
AI Infrastructure Work (Python Engineer) · Independent / ProjectsJan 2022 – Present · 4 yrs 11 mos
General Manager and Operations Lead · BMRS Corp and Pangaea & MedicineJan 2000 – Jan 2018 · 18 yrs 1 mo

Education

Bachelor's degree · University of Southern California (USC)2030 – 2026

Certifications and awards· 7

AWS Cloud Practitioner Essentials2026
Anthropic Claude (Model Context Protocol) - Advanced Topics2026
Anthropic Claude with the Anthropic API2026
Introduction to Model Context Protocol2026

Hire Edward Kim today

Post a job and invite Edward, or contact them directly. No hidden commission.

Post a job