Skip to content

The AI Automation Sprint

One production AI workflow, shipped live in 14 days.

A fixed-price sprint for SaaS, agencies, and professional services. $6,500 USD / €6,000 fixed. Support triage, lead enrichment, churn detection, onboarding personalisation, or content pipelines — your pick, my build, your team owns it on day 15.

No hourly billing. No scope creep. No demos. Live by day 14 or refunded.

What the Sprint ships · six options

Pick one. Ship it. Decide what's next from a position of data.

  1. 01

    Support Ticket Triage

    Every inbound ticket is classified, prioritised, and paired with an AI-drafted first response — before a human opens it.

    Best fit — SaaS support teams of 3+.

    Expected lift

    60–80% cut in first-response time · 30–50% ticket deflection

  2. 02

    Lead Enrichment + Personalised Outreach

    Inbound leads enriched with firmographics, intent signals, and LinkedIn activity — then scored, routed, and paired with a personalised opener.

    Best fit — B2B sales teams with 50+ leads/week.

    Expected lift

    3–5× lift in reply rate · 30–60% lift in lead-to-meeting

  3. 03

    Churn Signal Detection

    Weekly report surfacing at-risk accounts before the renewal conversation. Signals across product usage, support sentiment, and billing.

    Best fit — SaaS with product analytics in place.

    Expected lift

    12–20% reduction in mid-market churn

  4. 04

    Onboarding Personalisation

    Product-usage-aware email sequences that adapt to what each trial user has actually done — not a generic drip.

    Best fit — Self-serve SaaS with free trials.

    Expected lift

    15–35% lift in trial-to-paid conversion

  5. 05

    Content Pipeline

    End-to-end content workflow — topic monitoring, brief generation, AI drafting with brand voice, editorial review, scheduling.

    Best fit — Agencies, media, and content-driven brands.

    Expected lift

    One operator produces the output of a 5-person team

  6. 06

    Internal Reporting + Anomaly Detection

    Automated weekly metrics report with anomaly flags and suggested root causes. Replaces the Friday metrics meeting nobody enjoys.

    Best fit — Ops-heavy teams with metrics across 3+ sources.

    Expected lift

    Saves 4–6 exec hours per week

Not sure which fits? The free scoping call maps your highest-ROI candidate in 30 minutes. Or read the full playbook first.

The 14-day plan

Same rhythm every Sprint.

Day 0

Scoping call

Free 30 minutes. We pick the workflow, confirm the fixed scope, and agree the 14-day plan. You only commit after the call.

Days 1–3

Discovery and design

Deep dive with your process owner. Map the workflow on paper. Identify edge cases. Design LLM prompts and validation logic. You approve the blueprint before any code ships.

Days 4–9

Build

Wire up the workflow in n8n or Make. Integrate with your existing stack (CRM, help desk, data sources). Logging, monitoring, and error paths built in from day one.

Days 10–12

Shadow mode

Workflow runs alongside your existing manual process. Outputs compared side-by-side. Prompts tuned based on real inputs. Edge cases caught before users see them.

Day 13

Cutover and training

Live in production. Team trained on the dashboard, documentation delivered, monitoring verified. Ops team owns it from this point.

Day 14

Handoff

30-minute handoff call. Full documentation. 30-day support window opens. You own the workflow, you own the code, you own the data.

Four ways in

Start small. Scale from a position of data.

  1. 3 days · live

    Pilot Workflow

    $990USD

    €900 · £790 · A$1,490

    One small workflow — one trigger, one LLM step, one action — shipped live. The quickest way to work together on real code.

    • 30-min scoping call
    • Production workflow, live
    • One or two integrations
    • Credited toward the Sprint
  2. 8–12 weeks · 3 workflows

    Multi-Workflow Program

    From $18,000USD

    €16,500 · £14,500 · A$27,000

    The full stack — support triage, onboarding, and churn detection shipped in the order that compounds.

    • Scoping workshop with leadership
    • Three workflows sequenced for ROI
    • Shared data infrastructure
    • 60-day support across all three
  3. Monthly · ongoing

    AI Automation Retainer

    From $3,500USD / month

    €3,200 · £2,800 · A$5,300 / mo

    For teams shipping multiple workflows per quarter. Ongoing build, tuning, and monitoring — not the CTO seat.

    • 10–20 hours per week of dev
    • Monthly architecture review
    • 48-hour response time
    • Cancel with 30 days notice

Is this for you?

Honest fit, both directions.

This is for you if

  • SaaS companies (10–200 employees) drowning in support triage, onboarding churn, or lead qualification
  • Professional services firms (agencies, consultancies, accountants, law firms) where senior hours are eaten by repetitive client-coordination tasks
  • E-commerce and DTC brands with clear ROI on abandoned-cart recovery, product description generation, or post-purchase workflows
  • Founders who want to see a shipped workflow in 14 days before committing to a longer engagement
  • Teams who have tried AI tools, not AI workflows, and noticed the difference

This is not for you if

  • You want a chatbot on your marketing site (that is a different project)
  • You want to "explore AI" without a specific process to automate (book a workshop instead)
  • Your process changes weekly and has no stable pattern (automation needs stable patterns)
  • You want to pay hourly and drag the scope (the Sprint is fixed-price by design)
  • Your data sensitivity requires on-prem deployment in a regulated industry without an existing Cloud-permitted workflow platform

Tools & stack

Battle-tested tools, tool-neutral approach.

Orchestration

n8n (self-hosted or Cloud) · Make (Integromat) · Custom Python / TypeScript

AI / LLMs

Claude API (default) · OpenAI API · Gemini · Open-source models when fit

Data & Storage

PostgreSQL · Supabase · Airtable · Google Sheets · Your warehouse

Communication

Webhooks · Slack API · Email APIs · Telegram · WhatsApp Business

CRM & Sales

HubSpot · Salesforce · Pipedrive · Notion · Custom CRMs

Support

Zendesk · Intercom · Help Scout · Freshdesk

Infrastructure

Docker · AWS · Vercel · Netlify · Cron jobs · Queues

FAQ

The questions that come up on every scoping call.

What is the Pilot Workflow for?

The Pilot is a $990 USD (€900 EUR) on-ramp that ships real code in 3 days — not a report, not a deck. One trigger, one LLM step, one action, one or two integrations — a production workflow you can keep running the same day. Examples: an inbound lead enrichment-and-scoring pipe, a daily competitor digest to Slack, an AI-drafted support auto-reply, or a meeting-notes-to-summary flow. If you decide to proceed with the Sprint within 30 days, the pilot fee is credited in full. For most buyers this is the cleanest way to see how I work on real code before committing to a full build.

What exactly do I get for $6,500 USD?

One production AI automation workflow, shipped live in 14 days, with full documentation and 30-day support included. You choose the workflow (support triage, lead enrichment, churn detection, onboarding personalisation, or content pipeline) during a free 30-minute scoping call. Fixed price, fixed timeline, no scope creep.

What if my workflow needs more than 14 days?

The Sprint is scoped specifically to a 14-day deliverable. If the first call reveals a workflow that genuinely needs more (rare, but happens with deep-regulatory or 10+ integration scopes), I quote a Multi-Workflow Program instead — typically $18,000–$32,000 USD for 3 workflows over 8–12 weeks. No surprises; scope is clear before you commit.

Do you work with global businesses?

Yes — the majority of clients are. Based in Portugal, remote-first, async-by-default, calls scheduled across timezones. Invoicing in EUR for EU clients, GBP for UK, USD for US and rest of world, AUD for Australia. VIES-validated VAT reverse charge on intra-EU B2B. No physical presence required.

What tools do you build with?

n8n (self-hosted or Cloud) and Make for workflow orchestration. Claude API as the default LLM (Anthropic), OpenAI and Gemini when the task fits. Custom Python or TypeScript where the logic demands it. I pick the tool that fits your workflow, not the other way around.

Do I need technical staff to maintain the workflow after the Sprint?

No. Every Sprint ships with a dashboard, written documentation, and a 30-minute handoff call with your team. Most clients run the workflow in-house after handoff. For evolving workflows that need regular tuning, the monthly retainer is available.

What happens after the Sprint ends?

You have three options: (1) run the workflow in-house with the documentation provided, (2) book another Sprint for a second workflow, or (3) move to the monthly retainer for ongoing tuning and new workflow development. Most clients start with one Sprint and decide from there.

Is there a guarantee?

Yes. If the workflow is not live and producing measurable output by day 14, you pay only for time genuinely worked on discovery, and the rest is refunded. In three years of Sprints, this has happened zero times — but the guarantee is real.

Can I see what a finished Sprint looks like before I commit?

Yes. On the free 30-minute scoping call, I walk through two previous Sprints with real workflows, real dashboards, and real metrics (client names anonymised unless they have given explicit permission). You leave the call with a clear picture of what day 14 looks like for your specific workflow.

Free download · 6 pages · PDF

10 AI workflows that actually save hours.

Real agentic workflows running in production — not prompt packs. Stacks, costs, and failure modes from projects that shipped.

No spam. You also get the Sunday note — unsubscribe in one click.

David Dacruz — senior Bitcoin and AI engineer

David Dacruz

Replies in ~24h

Next step

Ready to ship one workflow in 14 days?

Free 30-minute scoping call. We identify the highest-ROI workflow for your business and confirm the fixed scope. You only commit after the call.