Blog — 20 notes · page 2 of 3
Notes from the build.
Problems I've hit, hurdles I've solved — written down so you don't have to hit them too.
Retell AI and GoHighLevel in Production: Dynamic Variables, Transfers, CRM Sync, and Retry-Safe Webhooks
A hands-on architecture for connecting Retell AI voice agents to GoHighLevel through n8n, Supabase, and custom code without duplicate contacts, stale context, unsafe retries, or unreliable appointment writes.
How to Build a Production AI Automation With n8n, OpenAI, Supabase, and Human Approval
Build a reliable n8n AI automation with OpenAI, Supabase, tools, memory, human approval, retries, observability, security, and production deployment patterns.
Scaling n8n Beyond One Container: Queue Mode, Redis, Postgres, Idempotency, and Burst Webhooks
A production architecture for scaling self-hosted n8n with queue mode, workers, Redis, Postgres, durable webhook ingress, idempotency, binary storage, backpressure, and tenant-aware load control.
RAG Is Not Dead: How to Build an Agentic RAG System With Hybrid Search, Reranking, Memory, and Evaluations
Learn how production RAG evolves into agentic retrieval with hybrid search, query planning, reranking, citations, memory, security, evaluation, and cost-aware orchestration.
HighLevel SaaS V1 vs V2 in 2026: The Architecture That Survives Custom Provisioning
A production architecture guide for choosing HighLevel SaaS V1 or V2, provisioning sub-accounts exactly once, controlling billing state, and adding n8n or Node.js where the native flow stops being enough.
How to Secure AI Agents Against Prompt Injection, Data Leakage, Tool Abuse, and Privilege Escalation
A practical security guide for AI agents covering threat modeling, prompt injection, tool permissions, MCP risks, data leakage, sandboxing, approvals, red teaming, and incident response.
When Zapier or Make Hit the Reliability Ceiling: Build a Durable Webhook Inbox
A production pattern for keeping Zapier or Make as the visible orchestration layer while moving webhook durability, deduplication, ordering, retries, and replay into Postgres or Supabase.
LLM Evaluations and Observability: How to Measure Whether Your AI Application Actually Works
Learn how to evaluate and monitor LLM applications with datasets, deterministic checks, human review, LLM judges, trajectory scoring, tracing, production metrics, and regression testing.