apiVersion: automation/v1
kind: Engineer
metadata:
name: Lorenz Leslie Espinosa
role: Automation Architect & IT Specialist
spec:
orgs:
- name: Leppard Law
focus: Legal ops — intake, case management, billing, client comms
- name: Le Cordon Bleu International
focus: IT infrastructure & marketing automation across multi-tenant Microsoft Cloud
architecture:
pattern: webhook-driven, API-first
buffer_layer: Airtable (staging between systems)
error_handling: retry 3x with backoff → fallback path → audit log → alert
principle: never fail silently
ai_agents:
production: LLOpenClaw v5.0 — multi-agent Slack bot for legal ops
framework: OpenClaw (orchestrator + specialist agents with human-in-the-loop)
models: Claude Sonnet 4 (primary) + free models (heartbeat, compaction)
tools: 21 MCP tools via n8n — zero tokens on deterministic tasks
exploring: [LangGraph, RAG pipelines, Temporal workflows, self-healing systems]|
|
Not workflows — production systems with error handling, observability, and measurable business outcomes.
More systems I've built
| System | What It Does | Impact |
|---|---|---|
| Multi-Agent Task Delegator | Central AI controller routes natural-language commands to specialized workflows | Routine task time -30% |
| Smart Contract Monitor | Real-time on-chain Ethereum event detection with anomaly alerting | Diagnosis time -60% |
| AWS Health Sentinel | Serverless monitoring of AWS Health events → Slack/email alerts | Incident response -40% |
| Discord-Jira Issue Bridge | Community issues auto-flow into project tracking | 100+ issues/month, zero manual entry |
| Community Security System | Automated phishing, scam bot, impersonation detection | Member retention +25% |
| Martech Data Pipeline | Unified 1.5TB event data across 6 clients for campaigns | Email CTR +30%, churn -18% |
| LLM Compliance Monitor | AI-RMF aligned pipeline monitoring 100+ agents for bias/drift | 100% regulatory approval (EU + US) |
|
Static analysis for n8n workflow JSON. Catches credential leaks, deprecated nodes, orphaned connections, and best-practice violations. npx n8n-lint . # lint all workflows
npx n8n-lint --fix workflows/ # auto-fix safe issues
npx n8n-lint --format json . # JSON output for CI16 rules · configurable · auto-fix · CI-ready |
Docker stack: AI agent gateway + workflow automation. OpenClaw + n8n + PostgreSQL in one git clone ...openclaw-n8n-stack.git
cp .env.example .env && docker compose up -dsaves ~$65/mo · MCP · multi-agent · Docker |
|
Retry with backoff, dead-letter queue, fallback paths, PII-masked audit logging. |
Client intake, missed call recovery, billing sync, AI case routing. |
Multi-agent orchestrator with confidence scoring and human-in-the-loop. |
|
Production deployment guide — identity stack, security, n8n MCP. |
Python toolkit — DLQ, audit logging, PII masking, staging sync. |
OpenPhone + n8n — missed calls, SMS, call logging. TCPA compliant. |
Every system I build follows this pattern. Every path terminates in an audit log. Nothing fails silently.
flowchart TD
A[/"Trigger<br/>webhook · schedule · event"/] --> B{"Validate<br/>required fields · auth · format"}
B -->|valid| C["Process<br/>transform · enrich · route"]
B -->|invalid| D["Reject + Log<br/>reason · timestamp · source"]
C --> E["Output<br/>API write · notification · record"]
C -->|failure| F["Retry 3x<br/>exponential backoff"]
F -->|still failing| G["Fallback Path<br/>dead-letter · manual queue"]
F -->|recovered| E
E --> H[("Audit Log<br/>Airtable · Slack<br/>PII masked")]
G --> H
D --> H
style A fill:#1a1b26,stroke:#7aa2f7,color:#fff
style B fill:#161b22,stroke:#7aa2f7,color:#c9d1d9
style C fill:#161b22,stroke:#7dcfff,color:#c9d1d9
style D fill:#7f1d1d,stroke:#ef4444,color:#fff
style E fill:#14532d,stroke:#9ece6a,color:#fff
style F fill:#78350f,stroke:#fbbf24,color:#fff
style G fill:#7f1d1d,stroke:#ef4444,color:#fff
style H fill:#161b22,stroke:#8b949e,color:#8b949e
|
Orchestration & Automation |
AI Agents & LLM |
Legal Ops & CRM |
"I don't automate tasks — I architect systems that eliminate them."
Building legal ops automation at Leppard Law · Orlando, FL

