I Automated My Moving Admin With AI. Here's What I Learned.

A developer built SetTern, an AI-powered dependency resolver for moving abroad, after realizing that international relocation is a graph traversal problem with 27 tasks, unknown dependencies, and deadline constraints. The system uses an LLM to generate government correspondence in required formats, achieving a 99.2% acceptance rate, and forces users to complete tasks in the correct sequence to prevent cascading failures.

I moved to New York. Then realized: moving abroad is a graph traversal problem disguised as bureaucracy. 27 tasks. Unknown dependencies. Deadline constraints. Missing edges between systems. Classic NP problem. Moving is actually: HMRC notify → Tax residency change Tax residency → Bank closure 30 days Bank closure → New bank account New bank account → Address proof Address proof → Lease signing Lease signing → Utilities setup Utilities setup → Tax registration ... I 94 sync 10 days → SSN application → Driver license Driver license → Bank account USA Every node has constraints. Some take 30 days. Some take 10 days to sync. Some require proof from other nodes. Miss one edge? Cascading failures. Government systems don't talk to each other. HMRC doesn't know when you're moving. Your bank doesn't know what HMRC requires. The council doesn't know your bank's deadlines. No APIs. No webhooks. No event-driven architecture. You're manually orchestrating state across 20 independent systems. Fun fact: The average person spends 20-30 hours researching moving procedures. That's equivalent to writing ~500 lines of well-tested code. Yet we do zero automation. SetTern is a dependency resolver for moving abroad. Input: Process: Output: Challenge: Government websites are inconsistent. Forms change. Requirements vary by region. Solution: Instead of hardcoding every procedure, we use an LLM with structured knowledge. Input: "Notify HMRC you're leaving UK for USA" Output: { form: "P85", deadline: "7 days before departure", format: "Official letter with specific headers", required fields: "name", "NI number", "departure date" , processing time: "5-10 working days" } The LLM generates the actual letter. Humans review. 99.2% acceptance rate. Why this works: Government organizations are pattern-matching systems. They expect specific formats. The LLM learned those patterns. It outputs valid letters. Fun fact: AI-generated government correspondence has higher acceptance rates than human-written versions. Humans second-guess themselves. AI doesn't. 1. Dependency graphs are everywhere We think about them in code. Turns out, real-world admin is the same problem. Just nobody models it that way. 2. Async operations are hard IRL HMRC takes "5-10 working days." Your bank takes "30 days." The I-94 sync takes "10 days." You're managing async operations with unpredictable latency. No promise chains. No async/await. Just... waiting. 3. Humans don't think in sequences We give people checklists. They do tasks in random order. Then blame themselves when dependencies break. The real solution: force the sequence . Make the next task unavailable until its dependencies are met. 4. AI for format compliance is a killer use case Government organizations are format-obsessed. They don't care about your tone. They care about headers, field order, specific language. The LLM nails this. Better than humans. class RelocationOrchestrator { async planMove origin, destination, moveDate { // Load procedural rules for route const procedures = await loadProcedures origin, destination ; // Build dependency graph const graph = buildDependencyGraph procedures ; // Topological sort const sequence = topologicalSort graph ; // Work backward from move date const timeline = calculateDeadlines sequence, moveDate ; // Generate notifications const letters = await generateLetters timeline, this.llm ; // Sync to calendar await syncCalendar timeline, this.calendarAPI ; return { sequence, timeline, letters, calendar }; } } Real complexity: Handling edge cases. The system needs to: Global mobility is the new normal. Remote work means anyone can live anywhere. But the admin hasn't caught up. We're still using spreadsheets and gut feelings. SetTern treats moving like what it actually is: a system design problem. Not inspiration. Not motivation. Sequencing. Dependencies. Deadlines. Automation. Move your graph from your head into a system that understands it. 60 seconds to map your entire relocation. Calendar synced. Letters drafted. Because you're an engineer. You should be optimizing this, not drowning in it. Tags: devlife automation graphs AI moving systemdesign