Services
AI Workflow Consulting
We find the parts of client work worth automating and build AI workflows for them around the tools you already use. Each workflow is tested on real examples, then handed over to the people who run it.
What we do
Built around how your firm already works
An AI workflow is a series of steps in your work where an AI model does part of the job, such as reading a document, drafting a reply or pulling out the details that matter. The rest runs in the tools you already use, and a person checks the output wherever a mistake would be costly. Every engagement starts with how the work gets done today: who gathers what, where it gets drafted, and where a person needs to check it. We look for the steps that repeat, and the places where a first draft, a summary or a quick answer would help.
Then we build the workflow around the tools you already use. That usually means your email, documents, spreadsheets, task tracker and practice software. Someone on your team reviews the output wherever a mistake would be costly.
Use cases
Workflows we build
These are common starting points. If the work you have in mind isn't here, describe it in your inquiry.
Proposal preparation
Drafts a first proposal from your intake notes, past proposals and standard scope language, ready for you to edit and send.
Meeting follow-through
Takes the decisions from a client meeting, files each task in the task tool your team already uses with the right person assigned, and drafts the follow-up email for someone on your team to send.
Recurring client reports
Gathers that month's or quarter's numbers and notes and drafts the report in your format, so your reviewer checks it instead of assembling it.
Answers from your own documents
Lets staff ask questions and get answers drawn from your own documents and past work, with a link to the source behind each answer.
Client intake
Turns an intake form or call notes into a structured client record and a list of what's still missing, ready for someone on your team to check.
Document checks
Checks incoming documents against your checklist and flags anything missing or inconsistent, for your reviewer to confirm.
New client kickoff
When a proposal is signed, sets up the project, folders, tasks and welcome email from your templates.
Client inbox triage
Sorts incoming client email, drafts replies from your past answers and flags anything that looks urgent for the right person to handle.
Research briefs
Gathers what's public about a prospect or a topic and drafts a sourced brief before a call.
How it starts
From inquiry to written quote
Every engagement is scoped and quoted individually. This is how we get there.
Inquiry
Tell us about the workflow or problem through the contact form or by email. A few sentences is enough.
Email follow-up
We read your inquiry and reply by email within two business days to arrange a call.
Free 20-minute fit call
We talk through how the work gets done today and what you'd like to change, and decide together whether it's worth scoping. Budget is part of the conversation.
Written scope and quote
If the work is a good fit, we send a written scope and quote. Work starts once the scope is agreed in writing, with discovery as its first step.
Discovery is included in every engagement. When the work is unusually complex, we may propose a separately priced discovery engagement first, so the quote for the full work rests on a close look at how it really gets done.
Scope
Clear boundaries, agreed in writing
An engagement covers what the written scope describes: the workflows we build, the tools they connect to, the information they may use and the requirements they have to meet.
Larger or follow-on work is welcome. Anything beyond the agreed scope is scoped and quoted on its own, so you always know what you're agreeing to.
You set the limits on your data. We agree up front what information each workflow needs and which outside software or AI providers may handle it, and testing stays within those limits.
Once work starts
What an engagement looks like
Discovery
We look at how the work gets done today and decide what's worth changing first.
Scoped implementation
We build the workflow around the tools you already use, with a person reviewing the output wherever a mistake would be costly.
Validation
We test each workflow on real examples and refine it until it meets the agreed requirements.
Training and handover
We document the workflow and train the people who will run it, then hand it over.
After handover
What's included, and what's optional
Every workflow is documented and handed over to the people who run it. Day to day, running a workflow means reviewing what it produces and approving or correcting it, and we train the people who will do that. Every engagement includes a check-in once the workflow has been in real use. After that, if you'd like help as your work or tools change, we can keep adjusting it. Ongoing support is optional and quoted separately.
Questions
Before you get in touch
An AI workflow is a series of steps in your work where an AI model does part of the job, such as reading a document, drafting a reply or pulling out the details that matter. The rest runs in the tools you already use, and a person checks the output wherever a mistake would be costly.
Every engagement is quoted individually. After the fit call, if the work is a good fit, we send a written scope and quote. Scopes vary too much from firm to firm for a range to be useful before we've talked. A budget figure helps us plan the fit call. If it won't cover what you have in mind, we'll say so on the call and suggest what it can cover.
A firm where the owners run the work: consultancies, agencies and professional practices, whether your clients are businesses or individuals. It's a description, not an eligibility rule, and headcount isn't the test. If your work follows the same pattern again and again, tell us about it.
It's a free 20-minute call, arranged by email after we read your inquiry. We talk through how the work gets done today and what you'd like to change, and decide whether it's worth scoping.
We build around the tools you already use wherever we can. If a workflow needs new outside software or an AI provider, we agree on it with you before anything is built.
We sign a mutual NDA before discovery. Only the people working on your engagement, and the AI providers we agree on with you, handle your material. During discovery we agree what information the workflow needs and which outside software or AI providers may handle it, and testing stays within those limits. By default, workflows send text to Anthropic and OpenAI through their paid business services, under business terms that don't use your inputs or outputs to train their models. Any other provider is agreed with you before it's used. We keep your material while we work together and delete it from our systems 12 months after your last engagement. Please keep confidential details out of the inquiry itself.
Yes. ZeroFault Labs is a US company based in the San Francisco Bay Area, and we're glad to hear from firms anywhere.
Start with a short inquiry
A few sentences is enough, and you don't need a written brief to start. Not sure what to automate? Tell us about the work that takes up your team's week and we'll help you pick the first thing worth automating.