Claude Code Claude MCP Agentic AI Terminal Full-Stack

Companies in the candid network want engineers who use Claude Code as a core part of how they build, not as a curiosity. You understand agentic workflows, you know how to configure and steer them, and you can ship real features through the terminal faster than most teams can through traditional development.

What you do

→ Use Claude Code in agentic and interactive modes to build, test, and refactor production code
→ Configure project context with CLAUDE.md files and custom commands
→ Work with MCP servers to extend Claude Code's capabilities into your toolchain
→ Review and steer agent output: knowing when to trust it and when to intervene
→ Use sub-agents, parallel tool calls, and multi-file edits effectively

Who you are

→ You use Claude Code daily: it's your default way of writing software
→ You understand how context windows, tool use, and agentic loops actually work
→ You're comfortable in the terminal and can work across multiple languages and frameworks
→ You know how to write good prompts, good CLAUDE.md files, and good commit messages
→ You can tell when the agent is going off-track and course-correct before it wastes tokens
→ You have a good understanding of LLM models, their strengths, weaknesses, and where they fall down
→ You're genuinely obsessed with agentic coding - you follow the subreddits, the Twitter accounts, the changelogs. You're not catching up, you're already there
→ You've shipped (probably far too many) projects - both side projects and real, working production software

Who you're not

× Still on the fence about whether agentic coding tools are worth using
× Waiting for AI to 'mature' before you integrate it into your workflow
× Someone who's tried Claude Code once or twice but hasn't built anything real with it

Apply for this role

Claude Code Specialist

What you do

Who you are

Who you're not

Apply to join the network.