Architect Lab
“Design systems. Own the stack.”
Turn a production‑grade chatbot into a fully‑controlled, enterprise‑grade AI platform, then graduate to the Marshal Lab for organization‑wide AI governance and policy.
Resource | Description | Action |
|---|---|---|
▶️ Kick-off Video | Vision framing, ownership goals, success‑metric definitions | |
📄 Architecture Diagram PDF | High‑level diagram of VM → Coolify → Docker services (Flowise, Weaviate, n8n, FastAPI/Node, NocoDB, Grafana/Loki/Tempo) | |
✅ Metrics Checklist | Uptime > 99.9 %, MTTD < 30 s, secret‑rotation ≤ 30 days, cost‑vs‑cloud baseline |
Resource | Description | Action |
|---|---|---|
▶️ Tooling Overview Video | Walk‑through of Coolify, Vault, Grafana + Loki + Tempo and how they replace the Engineer‑track SaaS services | |
📂 Starter Repo | Docker‑Compose files for all services, pre‑configured for Coolify import | |
📄 Setup Guide | Step‑by‑step VM provisioning (Hetzner, Hostinger, GCP, AWS), Coolify install, Vault init, TLS cert generation |
Resource | Description | Action |
|---|---|---|
▶️ RAG Chatbot Walkthrough | Deploy full stack, add mutual‑TLS, enhanced fallback (5× Retry → Circuit‑Breaker → Graceful‑Degrade) | |
📂 Full-stack Compose |
| |
📂 FastAPI Email Service |
| |
📂 Observability Pack | Prometheus exporter snippets + Grafana dashboard JSON for fallback & email metrics | |
📄 Security Hardening Guide | Mutual TLS setup, Vault‑managed API keys, OIDC RBAC for Grafana, secret rotation workflow |
Resource | Description | Action |
|---|---|---|
▶️ Automation Overview | Build a lead‑capture pipeline (Tally → n8n → NocoDB) with deduplication, dead‑letter queue and enrichment micro‑services | |
📂 n8n Workflow Template | Full workflow: Tally webhook → Validate → Deduplication → Enrichment → NocoDB insert → Slack/email alert | |
📂 Deduplication Service | Node.js | |
📂 Enrichment Service | FastAPI service calling Clearbit (or similar) and emitting OpenTelemetry spans to Tempo | |
📄 Run Book | Operational guide covering webhook validation, rate‑limiting, dead‑letter handling, alert routing and manual retry procedures |
Resource | Description | Action |
|---|---|---|
▶️ Polish & Deploy Video | UI polishing tips, final security checklist, one‑click Coolify deployment to production | |
📄 CI/CD Blueprint | GitHub Actions workflow: lint Flowise JSON, unit‑test FastAPI/Node services, build Docker images, push to private registry, trigger Coolify staging deploy | |
📂 Secret Rotation Guide | Vault cron job that rotates API keys & DB passwords every 30 days, automatic injection into running containers | |
📂 Alert Config Snippets | Publicly accessible demo of the agentic RAG chatbot and the end‑to‑end automation workflow | |
🌐 Live Demo Link | Publicly accessible demo of the enterprise‑grade RAG chatbot and the end‑to‑end automation pipeline |
Resource | Description | Action |
|---|---|---|
▶️ Teaser Video | Continue to the Marshal track for centralized policy enforcement for data residency, model usage quotas and ethical guardrails. |
Move beyond drag‑and‑drop and run the same Flowise / n8n pipelines on a self‑hosted, zero‑trust stack you control.
Keep writing FastAPI or Node micro‑services while Coolify orchestrates everything side‑by‑side with no‑code nodes.
Provision, roll‑back and monitor every component yourself – Coolify, Vault, Grafana + Loki + Tempo, mutual‑TLS, automated secret rotation.
Own the entire AI service in‑house, with enterprise‑grade SLAs, compliance reporting and predictable cost scaling.
Level Up - Ready to bring AI governance, policy enforcement and organization‑wide AI strategy? Jump to the Marshal Lab for AI policy engine, data‑residency controls and enterprise‑scale governance.