SWE (Terminal and CLI Dev Tools Focused)

SWE (Terminal and CLI Dev Tools Focused) [$75-$80/hr] Experienced Software Engineers to evaluate and compare the performance of AI-powered CLI coding agents on real-world infrastructure debugging tasks You will solve TerminalBench tasks: real-world broken infrastructure scenarios running inside Docker containers. You'll use AI CLI agents to help you. Each task presents a failing system (databases, networking, security, pipelines) that you must diagnose and fix by writing a bash script, guided by AI agents in turn. Role Responsibilities • Solve the same infrastructure debugging task with CLI-based AI coding agent • Diagnose broken systems inside Docker containers (databases, TLS, pipelines, replication, access control) • Write bash scripts that fix the root cause and survive service restarts • Compare agents' approaches and rank their performance after each task Good Candidature • 3+ years of experience in software engineering, with hands-on debugging of systems and infrastructure • Strong bash/shell scripting proficiency: you'll be writing non-trivial fix scripts from scratch • Docker and containerization experience: every task runs inside a Docker container you'll need to explore via docker exec • Infrastructure and systems debugging skills: experience with PostgreSQL, MySQL, Redis, nginx, TLS, systemd, log analysis, or similar • Familiarity with version control workflows (Git, PRs, issue tracking) • Experience with AI coding tools (Copilot, Cursor, Claude, or similar) is a plus: you need to effectively prompt and evaluate AI output, not just code yourself Project Timeline • Start Date: Immediate • Duration: 1-2 weeks • Commitment: Part-time (15-25 hours/week, with flexibility up to 40 hours/week)