AI solutionsAI agents and automation for recurring work

We connect language models such as Claude and GPT to your existing systems. Quotes, tickets and data entry then largely take care of themselves. Where it matters, an employee approves the result first.

AI and machine learning

We use AI where it saves work: agents that carry out tasks, automation, and assistants that answer from your own documents. An employee approves the important steps first.

01

AI agents and automation

Agents that carry out tasks in your own systems. Where it matters, an employee approves the result first.

  • + Building and deploying AI agents
  • + Workflow automation
  • + Tool and MCP integrations
  • + RAG on your own documents
  • + Approval by an employee
  • + Monitoring and cost control
02

Machine learning

Models that make predictions or find patterns in your data

  • + Predictive analysis
  • + Classification
  • + Regression analysis
  • + Time series forecasts
  • + Recommendations
  • + Anomaly detection
03

Language processing

Text analysis, chatbots and document processing

  • + Chatbots
  • + Sentiment analysis
  • + Text classification
  • + Translation
  • + Extracting names and data from text
  • + Document summaries
04

Image recognition

Analysing photos and video to automate work

  • + Object detection
  • + Image classification
  • + Face recognition
  • + OCR and document scanning
  • + Quality control
  • + Video analysis

How we work with AI

  1. 01

    Custom models

    If an existing model is not enough, we train or fine-tune one on your own data.

  2. 02

    Connected to your data

    The AI works with the systems and databases you already have.

  3. 03

    Privacy and transparency

    We document which data the AI uses, what it is allowed to do and how it reaches an outcome.

How an AI project runs

  1. 01

    Review the data

    1-2 weeks

    We check the quality of your data, define what is needed and whether it is feasible.

  2. 02

    Build the model

    3-6 weeks

    Preparing the data, choosing or training a model and testing the results.

  3. 03

    Connect and test

    2-4 weeks

    Building an API, connecting it to your systems and testing with real volumes.

  4. 04

    Go-live and monitoring

    1-2 weeks

    Going live, setting up monitoring and improving based on real use.

Average development time: 8-16 weeks

What we build with AI

Applications for concrete, recurring work

01

Chatbots and knowledge bases

An assistant for customer service or internal use that answers from your own documents.

  • + Answers with sources (RAG)
  • + Multilingual
  • + Handover to a human
  • + CRM integration
02

Predictive analytics

Forecast demand, revenue or churn from your own data.

  • + Revenue forecasts
  • + Churn prediction
  • + Demand planning
  • + Risk assessment
03

Image recognition

Automatic analysis of images and video, for example for quality control.

  • + Object detection
  • + Image classification
  • + OCR and document processing
  • + Defect detection
04

Document processing

Pull data from invoices, forms and contracts automatically.

  • + Data extraction
  • + Document classification
  • + Summaries
  • + Contract review
05

Speech to text

Turn conversations and audio into text and summaries.

  • + Transcription
  • + Voice commands
  • + Meeting summaries
  • + Customer call analysis
06

AI agents and workflows

An agent carries out a task from start to finish, with an approval step where needed.

  • + AI agents
  • + Automated decision rules
  • + Workflows across CRM, ERP and email
  • + Email processing

How we approach an AI project

Start small, test, then expand

  1. 01

    Exploration

    We pick a task, look at the available data and work out what automating it saves.

  2. 02

    Data preparation

    Collecting, cleaning and where needed labelling documents and data.

  3. 03

    Build

    Choosing a model, building the logic and testing on real examples from your business.

  4. 04

    Integration

    Connections to your systems through an API or MCP, plus a screen for your staff.

  5. 05

    Monitoring

    Tracking results and costs, and adjusting where needed.

Applications by industry

Examples of where AI can save work in a business.

01 — Healthcare

Medical diagnosis assistant

AI system that supports doctors by analysing medical images.

02 — Retail

Personal product recommendations

Machine learning that recommends products based on customer behaviour.

03 — Manufacturing

Quality control with image recognition

Computer vision that checks products automatically during production.

04 — Finance

Fraud detection

Machine learning that flags suspicious transactions immediately.

Technology we use

Language models, frameworks and tools

Frameworks
  • TensorFlow
  • PyTorch
  • Scikit-learn
  • Keras
  • OpenCV
Cloud AI
  • Azure AI
  • AWS SageMaker
  • Google AI Platform
  • OpenAI API
Languages
  • Python
  • R
  • JavaScript
  • Java
  • C++
Tools
  • Jupyter
  • MLflow
  • Weights & Biases
  • Docker
  • Kubernetes

Try the AI demo

Test our AI demos and see how an assistant answers questions and carries out tasks.

FAQ

Frequently asked questions

About AI solutions

What our clients say

“Our new website is really fast, with load times of 1 second, and everything responds intuitively. Exactly what we had in mind.”

Sanimax

Website · Haarlem
sanimax.nl
“A complete solution with AI integration, from app to platform. We are very happy with the end result.”

Heumin

AI app · iOS, Android & web
heumin.com
“Within a week our website was completely redesigned and all funnels were correctly integrated. We are very happy with the result.”

ABN Vloerverwarming

Website & funnels · Haarlem
abnvloerverwarming.nl

Need custom software built

Book a free call with our developers. We look at your situation and discuss what's possible for your web application or business software. There is no obligation.