AI Automation Engineer — Agentic Workflows for Recruitment (n8n / LLM APIs

About us We're Averon Advisory Partners, an AI automation agency that builds and integrates AI agents into recruitment agencies' existing systems. We take the admin that eats recruiters' days — resume screening, candidate sourcing, outreach, interview scheduling, inbox triage, ATS/CRM updates, compliance checks — and turn it into agentic workflows that run quietly in the background. We're early-stage and moving fast. We work in tight 1–2 week build cycles and ship real, working automations into client environments rather than slide decks. We're looking for a contractor who can build at that pace with us. What you'll be building You'll own end-to-end automation builds across client engagements, including workflows like: - Resume screening and shortlisting (with human-in-the-loop guardrails — no auto-rejects) - Candidate sourcing and enrichment - Personalised candidate/client outreach - Interview scheduling and coordination - Inbox triage and drafted replies - Pre-screen Q&A flows - ATS and CRM record updates - Reference and compliance checks with audit logging Our stack We're tool-agnostic and pick what fits the client. Our primary stack is n8n for orchestration and the Anthropic Claude API for the reasoning layer, but we work across the wider ecosystem and want someone who's comfortable doing the same: Orchestration: n8n (primary), plus Make, Zapier, Pipedream, or custom code (Python/Node) where it's the better fit LLM APIs: Anthropic Claude (primary), plus OpenAI, Google Gemini, and others — including frameworks like LangChain / LlamaIndex where useful Integrations: ATS/CRM platforms (Bullhorn, JobAdder, Vincere, etc.), email, calendars, and web data sources Plumbing: REST APIs, webhooks, OAuth, vector stores/RAG where relevant, with a strong eye on reliability and clean error handling You're a strong fit if you have: - Hands-on experience building real, deployed automations in n8n and/or comparable platforms (Make, Zapier, Pipedream, or custom code) - Solid experience with one or more LLM APIs — prompt design, structured outputs, tool/function calling; Claude experience is a plus but not required - Comfort with REST APIs, webhooks, OAuth, and JSON wrangling - A track record integrating with ATS/CRM or other SaaS platforms - A bias toward shipping — you scope tightly, build, test, and hand off without hand-holding Nice to have: - Recruitment / staffing domain knowledge - Experience with agent frameworks, RAG, or eval/testing patterns for LLM workflows - Experience designing human-in-the-loop and audit-logging patterns - Data-handling awareness around privacy (we operate under Australian Privacy Act obligations and treat compliance as non-negotiable) Engagement Project-based to start, with strong potential for ongoing work as our client base grows. Most builds run in 1–2 week sprints. Remote, async-friendly; we care about output, not hours logged. Budget: $600 - $2,000+ AUD. *This is an initial project. Successful candidates will have access to ongoing client implementation work. How to apply To keep this quick, please: Start your proposal with the word "Shortlist" so I know you read this. Share 1–2 specific automation workflows you've built — ideally something involving an LLM and an external integration. A short Loom walkthrough beats a paragraph. Tell me, in 2–3 sentences, how you'd approach a resume-screening workflow that flags top candidates but never auto-rejects anyone. I read every proposal that follows those three steps and reply fast.