DWG A-101 — AUTOMATION PORTFOLIO
Sheet 1 of 1 · rev c · 2026-10 · open for projects

Automation systems, not zaps.

Human-in-the-loop approvals where money or reputation is at stake, deterministic core logic, and failure handling designed in rather than bolted on. The repo runs on n8n with custom code where a native node isn't enough — Make, Zapier or another stack when a client's setup genuinely calls for it, including the jobs that really are one simple zap. Every workflow below is a real, documented system — import it, press Test, and watch it route every branch with zero credentials.

5
n8n systems
224
nodes total
0
creds to demo
CI
checked every push
MIT
license
S-01

Featured workflows

SCALE — 1 NODE : 1 NODE

Each ships with a problem statement, architecture diagram, setup path, input contract, and a demo lane exercised by CI on every commit. Not toy demos — the same structure delivered on paid work.

WF n8n/client-report-generator 87 nodes

Client Report Generator

Weekly per-client marketing reports
  • Meta Ads + Google Ads + GA4 behind one adapter contract
  • Config-sheet driven — new client = one spreadsheet row
  • Draft-review mode: Gmail draft instead of auto-send
PASS — pilot: 16 report items, exactly 1 digest, alerts 3/3 · retryOnFail 15/15 credentialed nodes
OPEN →FLAGSHIP
WF n8n/promise-ledger 60 nodes

Promise Ledger

Track promises made to you in email
  • AI extraction of commitments from inbound mail
  • Chase drafts held behind an approval gate
  • Fulfillment detection closes the loop
PASS — pilot: 5-item happy path, alerts 3/3 · retryOnFail 19/19 · error-alert lane included
OPEN →GMAIL · SHEETS · OPENAI
WF n8n/faq-chatbot 44 nodes

FAQ Chatbot

Support chatbot that grows its own FAQ
  • Learns from misses — only with owner approval
  • Lead capture built into the conversation flow
  • Error-alert lane wired for workflow-level failures
PASS — pilot: 6 visitor turns routed, digest exactly 1, 3/3 edge branches · retryOnFail 14/14
OPEN →CHAT · SHEETS · OPENAI
WF n8n/daily-cash-tally 20 nodes

Daily Cash Tally

End-of-day till reconciliation
  • 30-day history scoring per cashier
  • Repeat-offender escalation path
  • Quarantine lane isolates suspicious counts
PASS — demo exercises all 5 branch routes: quarantine, escalation, gap alert · retryOnFail 9/9 credentialed nodes
OPEN →SHEETS · GMAIL · TELEGRAM
WF n8n/affiliate-intake 13 nodes

Affiliate Intake

Screen affiliate applications
  • Editable scoring rules — no code changes
  • Outreach drafts for qualified applicants
  • Fully credential-free — import and run
PASS — demo: 6 applications → 2 qualified drafts + 4 reason-coded alerts, exactly 1 digest
OPEN →ZERO CREDENTIALS
S-02

Flagship detail — Client Report Generator

FIG. 1 · ASSEMBLY
Client Report Generator architecture diagram
FIG. 1 — 87-node assembly. Config sheet drives sources, sinks, thresholds; red lanes are human approval / exception paths.

Field notes

Weekly per-client marketing reports, end to end: three data sources behind one adapter contract, deterministic deltas, a human review seam, an idempotent send ledger.

  • AI writes prose around a finished table — it can never invent a number
  • Ledger keyed on client + ISO week — no double-sends on re-runs
  • One bad API doesn't kill the batch — failures route to a composed alert
  • Error lane armed via Error Workflow settings — documented, not implied

FULL CASE STUDY →

S-03

Who this is for

4 PROFILES

Different businesses, same disease — repetitive work silently eating hours. Each profile maps to a working system in the repo.

Agencies & service firms

Client work that scales without headcount

Weekly client reports, onboarding checklists, approval chasing — multiplied by every retainer you sign.
SHIPS AS › client-report-generator — config-sheet driven, draft-review mode
Solo founders & small teams

An ops layer you don't have to hire for

Leads going cold in the inbox, follow-ups you meant to send, promises you forgot were made to you.
SHIPS AS › promise-ledger + faq-chatbot — with approval gates
Retail & hospitality

Cash and shift operations that self-check

End-of-day tallies, drawer discrepancies, repeat-offender patterns nobody has time to cross-reference.
SHIPS AS › daily-cash-tally — 30-day scoring, quarantine lane
Partner & creator programs

Intake that filters before you read

Affiliate and partner applications piling up — most unqualified, all needing a polite, timely answer.
SHIPS AS › affiliate-intake — editable rules, zero credentials
S-04

Legend — how to read the work

REF: PAPER PALETTE
Engineering habits on display — every row is visible in the repo, not a claim.
MarkingMeaning
gate / approvalHuman-in-the-loop approval gates where money or reputation is at stake
retryOnFail · maxTries 3Retry/backoff on every credentialed node — transient failures absorbed
appendOrUpdate · dedupeIdempotent writes — re-runs are safe by construction
quarantine laneSuspicious records isolated instead of dropped or passed through
ai at the edgeDeterministic core, LLM at the edges — it drafts, it never decides
demo laneZero-credential test path — import, press Test, watch every branch
S-04a

Process

5 STATIONS
ST-1

Discovery

Map your process, find the automation boundary.

ST-2

Spec

You approve a written spec before anything is built.

ST-3

Build

Workflow + error handling + docs, shared workspace.

ST-4

Verify

Tested on sample data, then a real pilot run.

ST-5

Handoff

You own everything — workflows, docs, credentials.

S-05

Bill of materials — services

QTY AS REQUIRED
ItemPartSpecificationQty
BOM-00Automation auditSend the task eating your week — get a written teardown naming what to automate first, with rough effort on each. See S-06.free
BOM-01Workflow setupDesign, build, and deploy your workflow — n8n, Make, Zapier, or custom code; cloud or self-hosted.per workflow
BOM-02n8n self-hostingDocker/VPS deployment, queue mode, backups, monitoring.per instance
BOM-03Workflow audit & rescueFix brittle or broken workflows; add error handling and retries.per audit
BOM-04Document processingOCR + LLM extraction on invoices, POs, intake forms — structured into your systems, low-confidence routed to a human.per pipeline
BOM-05Custom integrationsConnect APIs that have no native nodes — Python/JS custom nodes.per connector
BOM-06AI agent pipelinesLLM-powered classification, drafting, enrichment inside your stack. AI drafts — it never decides alone.per pipeline
BOM-07Ongoing maintenanceMonitoring, alerting, versioned updates.per month
S-06

Free automation audit

ISSUED FOR HIRE
Open for work

Tell me the task that eats your week.

AVAILABLE FOR SHORT-TERM PROJECTS — WORKFLOW FIXES · NEW BUILDS · ERROR-HANDLING AUDITS

I'm Vinh Le — I design and build automation systems for small businesses, agencies, and solo founders. Send me 2–3 sentences about the process draining your time and the tools you run on. You get back a written teardown naming what I'd automate first, in order, with rough effort on each — free, by message, no call to sit through. If the honest answer is that a task shouldn't be automated, the audit says so.

What every engagement ships
  • Architecture diagram + written setup path — documented, not tribal knowledge
  • A demo lane you can run before buying: import, press Test, watch every branch — zero credentials
  • Failure handling built in — retries on credentialed nodes, error-alert lanes, quarantine instead of silent drops
  • Full handover — you own the workflow, the docs and the credentials. No lock-in