The workflow catalogue
Twenty workflows we have already built, named the way you search for them.
Every row is running in production today, for a client or inside one of our own products. Find the one that matches your problem and start on its page: it names what the workflow does, where it already runs, and what you receive.
All twenty workflows, by family
Research and intelligence.
Work that reads so your people decide: sources read overnight, records answering in plain English, archives turned into databases.
AI research agent
Two to six weeks, then a retainer
An agent that reads your sources overnight, works out what changed, and writes it up in plain English, with every figure computed in code rather than by a model.
Read the page
Intelligent document processing
Three to six weeks, fixed fee
Your PDFs, Word files and spreadsheets become one database anyone in the office can query, held in standard Postgres in your own account.
Read the page
Ask-your-data agent
Two to four weeks
A plain-English question box over your own records. The model turns the question into a query, the database does the arithmetic, and the answer arrives with the query that produced it.
Read the page
Customer-facing.
Work your customers meet: questions answered around the clock, enquiries qualified in minutes, threads that never go quiet.
AI chatbot for customer service
Three to six weeks, then a retainer
A chatbot built for businesses in Dubai and connected to your live records rather than a page of FAQs. It answers from the data, hands over to a person where you decide, and says it does not know instead of inventing.
Read the page
AI voice agent
Three to six weeks, then a retainer
A voice interface over your own systems. Someone asks a question out loud, gets a few sentences back, and can ask the agent to explain how it reached the answer.
Read the page
AI lead qualification agent
Three to five weeks, then a retainer
An agent that replies to each enquiry, asks the questions that decide whether it is worth pursuing, and scores the answers so your team works the best ones first.
Read the page
AI email agent
Three to five weeks, then a retainer
Inbound email wakes the agent. It reads the message, files it against the right contact and conversation, and answers it under rules you wrote down.
Read the page
AI follow-up automation
Two to four weeks
An agent that watches for what went quiet, and sends the follow-up at the interval you chose, in your tone, with a record of every nudge.
Read the page
Automated review requests
One to three weeks
An agent that asks a client for a review at the right moment, chosen by rules you set, and never asks the same person twice.
Read the page
Commercial.
Work that wins and prices the job: prospects scored daily, quotes computed by your real rules, spend capped in code.
AI quoting engine
Four to eight weeks
A public configurator that runs your real costing rules, so a customer designs the job, sees the price, and sends you an enquiry that is already a quote.
Read the page
AI outreach agent
Three to five weeks, then a retainer
An agent that scores your prospect list daily and drafts the pitches worth sending, with every number in them pulled from live data rather than remembered or invented.
Read the page
AI ad management agent
Four to six weeks, then a retainer
An agent that runs the campaign lifecycle end to end: audiences set up, campaigns launched, budget moved to what is working, finished campaigns closed out and archived.
Read the page
AI budget control agent
One to three weeks, fixed fee
A hard ceiling on automated spend, enforced in code before any money moves, with a scoreboard showing spend against the cap in real time.
Read the page
Operations.
Work that keeps the business informed: reports that assemble themselves, monitors that escalate to a person, paperwork generated from the record.
Automated reporting
Two to six weeks, then a retainer
An agent that collects from every source on a schedule, computes what changed, and sends the report in the format your team actually reads.
Read the page
AI monitoring agent
Two to five weeks, then a retainer
An agent that watches your feeds, numbers and systems continuously, and escalates to a named person when a threshold you set is crossed.
Read the page
Document generation automation
Two to four weeks, fixed fee
Automation that turns a record into the documents the business runs on: the prep sheet, the run sheet, the plan, generated from live data by rules.
Read the page
Content and publishing.
Work that publishes under control: written once, checked at a gate, distributed everywhere, and kept true after it ships.
AI blog writing agent
Three to six weeks, then a retainer
An agent that finds what your audience is searching for, writes the piece, and passes it through three gates before publishing: rules, an editor pass, and a fact check.
Read the page
Multi-channel publishing agent
Three to five weeks, then a retainer
An agent that takes one piece of writing and turns it into the blog post, the newsletter and the social threads, each shaped for its channel, all passing one safety gate before anything ships.
Read the page
AI quality control agent
Three to six weeks, then a retainer
An agent that audits your published content on a schedule, against a source of truth and rules you set, and repairs what drifts, with the repairs verified before they go live.
Read the page
AI message drafting
Two to four weeks
An agent that drafts your outgoing customer messages in bulk, each grounded in that customer's actual record, for a person to review and send.
Read the page
If your workflow is not on this list.
The list is what we have already built, not the limit of what we build. Describe the process, and the first call tells you which of these it is closest to, what it would take, and what it should return.
Discuss a business process
Thirty minutes on the process you want to improve. You will know what the right solution looks like, what it would take to build, and what it should return.
Replies come from a named person in Dubai within one working day.