Description
Summary We're an AI automation agency running workflow agents (n8n) for business clients. One of our existing builds has bugs and needs optimization. Looking for an engineer to step in, diagnose, and fix — not build from zero. The Work ● Review an existing n8n workflow agent (classifies incoming work, routes by rules, escalates urgent cases to a human) ● Fix bugs in classification accuracy, routing logic, and escalation triggers ● Improve error handling and reliability under real client data ● Optimize LLM calls for cost/latency where relevant ● Light documentation of what was changed and why Tools: n8n (main orchestration), OpenAI/Anthropic/Gemini APIs, integrations across Slack, HubSpot, GHL, Google Workspace as needed. Show us something you've built or fixed A short walkthrough (video or write-up) of a workflow automation you've debugged or improved — what was broken, what you changed, and the outcome. What matters most ● Comfortable reading and debugging someone else's existing workflow, not just building from scratch ● Fixes that hold up in production with real client data ● Clear communication on root cause and what changed Questions 1. Describe a time you debugged or fixed an existing n8n (or similar) automation you didn't originally build. What was wrong, and how did you find it? 2. How do you approach auditing an unfamiliar workflow before making changes? 3. What's your availability this week for a short, focused project? Proposals without a relevant example will not be reviewed.