AI automation
AI automation: for the work where a model has to understand text
Recurring work that calls for reading and understanding, and that a rule-based flow can't handle: summarising, parsing documents, classifying, generating content. We build the AI in where it demonstrably saves hours, and nowhere else.
- AI only where it saves time
- Fixed price agreed upfront
- Output with source citations
In short
AI automation is for the steps where a model genuinely has to understand something: summarising a ruling, parsing a CV, classifying an archive, generating a document. Ordinary, predictable flows where a model doesn't need to understand text (connecting systems, moving data, RPA) belong to our separate Automation service. This page is about the steps where AI truly has to understand. At BBS Advocaten, AI turns a court ruling into a blog: 5+ hours a week saved, 10x more content. At Simply, AI parses CVs where older OCR fails and every recruiter wins back four to six hours a week. At Planit, around 40% of the consultancy work is automated. Fixed price agreed upfront, and AI only where it demonstrably saves hours.
What gets in the way
Not everything you want to automate is a clean, predictable run of steps. Sometimes someone has to read and understand something before the next step can happen: summarise a ruling, take apart a CV, classify a document, write text in the house style. That's exactly the work a rule-based flow can't handle. And that's what AI automation is about. To be clear about the line: ordinary, predictable flows where a model doesn't need to understand text (connecting systems, moving data, RPA) belong to our separate Automation service. This page is about the steps where AI truly has to understand: summarising, parsing, classifying, generating. We don't bolt AI onto a process to make it sound modern. We use it where it demonstrably saves hours, and we're honest about when it isn't needed. At BBS Advocaten, an AI step turns a court ruling into a publishable blog and social post: 5+ hours a week saved and ten times more content. At Simply, AI document parsing reads in CVs and vacancies where older OCR fails, and every recruiter wins back four to six hours a week. At Planit Consulting, an AI pipeline makes an archive of 18,429 documents searchable with source citations, good for around 40% of the daily consultancy work. We think along with you: we listen first to what you want to achieve now and where you are heading, then choose the technology to match, for the most scalable and powerful result that is cost-efficient at the same time.
Recurring work that calls for reading and understanding
Some steps repeat every week, but you can't capture them in a fixed rule because something has to be read and understood first. Summarising a ruling, taking apart a CV, judging an email on intent. A plain flow stalls on that. At BBS Advocaten that exact manual work cost 30 to 45 minutes per ruling.
AI as a gimmick instead of real time saved
AI gets bolted onto processes everywhere to sound modern, while in practice it returns nothing. A chatbot nobody uses, a button that gives half an answer. We build AI only where it measurably saves hours. At Simply the AI took over the admin that cost recruiters four to six hours a week.
Output you can't trust
AI that occasionally makes something up is worse than no AI. If you can't check the answer, you won't dare rely on it. That's why we build in source citations and checkpoints. At Planit every answer points back to the original document, so consultants can always verify what the AI says.
Old tools that can't get the text out of your documents
Classic OCR trips over scans, odd layouts and complicated CVs, and produces junk someone still has to go through by hand. At Simply we built AI document parsing that reliably reads in CVs and vacancies where that older OCR fails, and pulls the data out cleanly.
What we build
How we help
AI summarisation pipelines: from source to publishable content
The pipeline reads sources, such as RSS feeds on categories and keywords, and an AI step turns the source document into a blog with title, header and summary plus a matching social post, all in the house style. Uploaded PDFs run through the same pipeline. Result: from source to publication in minutes, instead of hours of manual work.
AI document parsing where OCR fails
We build AI document parsing that reliably reads in documents like CVs, contracts and forms where older OCR trips up, pulls the data out and pushes it into your system. No more messy output someone has to correct by hand, but clean, structured data that's usable right away.
AI document creation in your own house style
AI document creation builds documents in your house style in seconds, from your own templates. Not one generic format, but documents that immediately look as if someone made them by hand in your style, ready to send.
Archive automation that makes things searchable with source citations
We build an AI pipeline that classifies your documents by type and tier, chunks them semantically and makes them searchable. So you find precedents in seconds instead of hours, with a citation back to the original document, so you can verify every answer.
AI only where it demonstrably saves time
We don't start with the model, we start with your process. Which step genuinely needs to understand text, and where does an ordinary, predictable flow do the job? We make that call honestly. We build the AI steps exactly where they save hours, with checkpoints and source citations so you can trust the output.
Proof
We have built this
Real projects we can name.
BBS Advocaten
Legal
A legal content pipeline that turns court RSS feeds into ready-to-publish blogs and social posts.
5+ hrs/week
Time saved
10x more content
Output
Simply
Recruitment SaaS
An AI recruitment suite, a native Salesforce app, a secure back-office and a marketing site. The full Simply stack, built end to end.
4-6 hrs/week
Recruiter time saved
Native Salesforce (Apex)
Integration
Planit Consulting
Planning consultancy
An AI chat that makes 18,429 planning documents searchable, with source citations.
18,429
Documents
95,592
Chunks
Our approach
From idea to working software in four phases
Every project runs through the same four phases, so you always know what's happening, what comes next and what it costs — from first call to live software, usually in weeks rather than months.
- 01 Phase 1 Free intake
Understand & analyse
Everything starts with a good conversation. We map your goals, processes and the bottlenecks worth solving — no sales pitch, just an honest read on where the biggest win sits.
- 02 Phase 2 Free blueprint
Blueprint & quote
We turn the analysis into a blueprint: a clear plan showing exactly which steps we'll take and why, paired with one fixed quote. You know up front precisely what we build, what it costs and when it's done — no open ends, no hourly billing, no surprises afterwards. And getting that blueprint is completely free, no obligation.
- 03 Phase 3 Prototype in days
Build, test & deploy
Then we build, in short iterations. You see a working prototype in days and a finished product in weeks. We work on a modern, AI-native and secure stack — with data protection and GDPR in mind from day one.
- 04 Phase 4 100% your code
Implement & optimise
We launch, hand over the full codebase and keep improving on your terms. You get 100% ownership of the code we write — no lock-in, no licensing games. Hosting and maintenance are optional, never required.
What you always get
Fixed price
Agreed up front, never an open end.
100% code ownership
Your code, fully yours, no lock-in.
Live in weeks
Prototypes in days, finished in weeks.
Modern & secure
A scalable stack that always integrates with the latest tech. Secured by our in-house cybersecurity experts.
Tools & tech
We're not tied to one model or platform. We pick the AI approach that fits your data, budget and goal — an off-the-shelf model, a custom pipeline, or a mix — and we're honest when simpler tech does the job just as well. No hype, no lock-in.
What you get
- A working AI automation that runs on your own environment and data
- AI steps only where understanding text is needed: summarising, parsing, classifying, generating
- Document parsing that reliably pulls text from CVs, PDFs and scans where OCR fails
- Content and documents that roll out of your data in your own house style
- Source citations and checkpoints, so you can verify and trust the output
- An honest call upfront: which step needs AI and which doesn't
- A fixed price agreed upfront, with no open ends
- Optionally maintenance and ongoing development when a source or process changes
Our promise
Your software stays yours
- Full code ownership The full source code is and stays yours, documented and ready to hand over.
- No vendor lock-in Stop working with us and your system keeps running. Any developer can pick it up.
- Fixed price up front You get a no-obligation blueprint and quote up front: no open-ended billing and no surprises afterwards.
- Security in every layer We build security-by-design, with a specialist in-house.
Frequently asked questions
What does it cost to have AI automation built?
That depends on the scope, but you always get a fixed price agreed upfront. No open-ended hourly billing. We map your process first, work out which step genuinely calls for understanding text, and then give a price for what you actually need. A single AI summarisation step is a matter of days to weeks; a full RAG pipeline over a large archive, like at Planit, costs more. Either way, you know where you stand from the start.
How long before an AI automation goes live?
A single AI step, like the summary at BBS, often lives within days to weeks. A larger pipeline takes longer: at Planit an archive of 18,429 documents first had to be classified, chunked and embedded before the chat could run. We work in short iterations, so you see a working version quickly. The exact timeline depends on how much data is involved and how clean it is, and we agree that upfront.
Does the AI automation stay mine, and who maintains it?
The automation runs on your own environment and the code and flows are yours, with documentation and a handover. No vendor lock-in: your own team or another party can continue with it. If you'd like us to maintain it, for instance because a model or source changes, we can, on clear terms, but you're never tied to us.
Will this AI replace my people?
No, it takes the dull reading work off their plate. The principle stays the same: people do the judgement work, the AI does the repetition. At BBS a lawyer still decides what's relevant and what the tone should be; the AI handles the first draft. At Planit, consultants still decide which precedent counts; the AI just finds it faster, with citations so they can read it back themselves. Your team is freed up for the work that actually matters.
Is AI really needed here, or could it work without?
Good question, and we ask it ourselves too. Plenty of automation needs no AI at all: connecting systems, moving data or taking over screen work is predictable work that belongs to our separate Automation service. We bring in AI only where a step genuinely has to understand text, like summarising a ruling or parsing a CV where OCR fails. If you can do without, we'll say so. You don't pay for AI when it adds nothing.
How do I know I can trust the AI output?
By making it checkable. We build in source citations where we can: at Planit every answer points back to the original document, so consultants can always verify where it came from. And where a step needs human judgement, we build in a checkpoint instead of a blind pass-through. That's how you win time without flying blind on something you can't review.
Ready to build?
Book a no-obligation call — we'd love to think through the best approach with you.
Book a call