Training course · Model Context Protocol (MCP) · Data Science & Machine Learning

Model Context Protocol (MCP) for Data Scientists and ML Engineers

Build and connect MCP servers securely: tools, resources and prompts for Claude Code, Cursor, Copilot and your own AI agents.

Duration
2 days (14 hrs)
Format
Online or on-site
Language
German or English
Participants
open dates or exclusive for your team
Certificate
Certificate of attendance

Overview

This course connects Model Context Protocol (MCP) with the everyday processes of Data Science & Machine Learning. Participants work on realistic scenarios and leave with prompts, workflows and checklists they can use straight away.

The Model Context Protocol is an open standard that gives AI applications a uniform way to access external tools, data sources and workflows. Instead of building separate integrations for every assistant, you implement an MCP server once and use it in Claude Code, Claude Desktop, Cursor, GitHub Copilot, VS Code and your own agents. In this two-day course you learn the host, client and server architecture, the three core primitives of tools, resources and prompts, and the stdio and Streamable HTTP transports, applying each of them directly in your own code.

On day one you use the official TypeScript or Python SDK to build an MCP server that exposes internal APIs, databases or files to AI assistants in a controlled way, and you test it with the MCP Inspector. Day two covers remote servers, OAuth-based authorisation, security risks such as prompt injection and tool poisoning, and operations, logging and governance in an enterprise setting. You learn to write tool descriptions that models use reliably, and you leave knowing how to integrate MCP servers safely into your development and agent landscape.

Data science and machine learning thrive on moving quickly between exploration, modelling and production. A great deal of time goes into data preparation with pandas or Polars, feature engineering, visualisation, hyperparameter experiments and finally turning notebook code into maintainable pipelines. AI coding assistants and agents speed up every one of these steps: they write exploration and plotting code, suggest validation strategies, refactor notebooks into tested Python modules, build scikit-learn or PyTorch training scripts and set up experiment tracking with MLflow. Methodological rigour is essential, though, because plausible-looking code can hide data leakage or misleading metrics. This course teaches you to use AI productively while safeguarding the statistical quality of your results.

The course is delivered live – either online in a virtual classroom or on-site in our training rooms in Munich, Salzburg, Berlin, Augsburg, Hanover, Bremen, Stuttgart and Kaiserslautern. We can also run it exclusively for your team.

Who is this course for?

  • Data scientists in business units and analytics teams
  • ML engineers responsible for training and deployment
  • Data analysts moving into Python and machine learning
  • Researchers and developers with notebook-heavy workflows
  • MLOps teams running models in production

Typical challenges we address

  • Messy notebooks that are hard to reproduce and put into production
  • Time-consuming data preparation and feature engineering
  • Methodological pitfalls such as data leakage, unsuitable metrics or flawed validation
  • Lack of traceability for experiments, models and data versions

What you will be able to do

  • You explain the MCP architecture and choose between tools, resources and prompts for each use case.
  • You develop your own MCP server in TypeScript or Python and test it systematically.
  • You connect MCP servers to Claude Code, Cursor, Copilot and custom agents.
  • You run remote servers using Streamable HTTP and OAuth-based authorisation.
  • Accelerate exploration, feature engineering and modelling with AI assistants without sacrificing methodological rigour.
  • Use AI agents to turn notebook code into tested, reproducible ML pipelines.
  • Recognise typical errors in generated ML code, such as data leakage or unsuitable metrics, and correct them deliberately.

Course outline

Day 1

Module 1 Architecture and core concepts
  • Roles of host, client and server and the lifecycle of a connection
  • JSON-RPC messages, capability negotiation and notifications
  • Tools, resources and prompts: choosing the right primitive
  • Client features such as sampling, roots and elicitation at a glance
Module 2 Building your own MCP server
  • Setting up a server with the official TypeScript or Python SDK
  • Defining tools cleanly with JSON Schema for inputs and outputs
  • Exposing resources and resource templates for documents, records and configuration
  • Testing and debugging with the MCP Inspector and meaningful error messages
Module 3 Connecting to Claude Code, Cursor, Copilot and agents
  • Configuring local stdio servers and remote servers in clients
  • Project-level and user-level configuration, approvals and permissions
  • Writing tool descriptions that models reliably select correctly
  • Integrating MCP servers into custom agents built with the Claude Agent SDK or OpenAI Agents SDK
Module 4 Transport, authorisation and operations
  • Streamable HTTP, session management and stateless servers
  • OAuth-based authorisation, token validation and integration with existing identity providers
  • Deploying as containers, on serverless platforms or behind a gateway
  • Logging, monitoring, rate limiting and versioning tool interfaces

Day 2

Module 5 Security and governance
  • Threats: prompt injection via tool results, tool poisoning, confused deputy
  • Least privilege, confirmation requirements and sandboxing
  • Reviewing and approving third-party servers within the organisation
  • Data protection, logging of personal data and requirements under GDPR and the EU AI Act
Module 6 Practice lab: Data Science & Machine Learning
  • Exploratory data analysis
  • Turning notebooks into pipelines
  • Model training and validation
  • Experiment tracking and MLOps
  • Explaining and auditing models
  • Evaluating LLM and RAG prototypes
Module 7 Data protection, quality & transfer to daily work
  • Training and analysis data often contains personal information, so use schemas, aggregates or synthetic data in prompts and observe GDPR principles such as purpose limitation. For high-risk applications, the EU AI Act sets requirements on data quality, documentation and transparency, and the licences of datasets, pre-trained models and Python packages must also be checked.
  • Ground rules for safe AI use in your team
  • Checking results: facts, sources, tests and the four-eyes principle
  • Personal action plan for the next 30 days
  • Open questions, outlook and resources for further learning

Practical scenarios from your field

Exploratory data analysis

The assistant writes code for data profiles, distributions, correlations and missing values with pandas or Polars, produces meaningful visualisations and summarises anomalies as hypotheses for modelling.

Turning notebooks into pipelines

The agent refactors notebook code into Python modules with functions, configuration and pytest tests, removes hidden state and makes training and inference reproducible. Results become reliably comparable.

Model training and validation

The AI builds scikit-learn pipelines or PyTorch training loops with proper cross-validation, suitable metrics and baselines, and actively flags risks such as data leakage or class imbalance.

Experiment tracking and MLOps

The assistant sets up MLflow or similar tools for parameters, metrics and artefacts, versions models and creates training and deployment workflows for the CI/CD pipeline.

Explaining and auditing models

Using SHAP, permutation importance or partial dependence plots, the AI produces explanations of model decisions as well as analyses of fairness and robustness across relevant subgroups.

Evaluating LLM and RAG prototypes

The AI helps build evaluation datasets, automated scoring metrics and comparison experiments for prompt variants, embedding models or retrieval strategies. Quality, cost and latency of variants can be compared objectively.

Possible hands-on projects

Churn prediction with explanations

A complete project from data preparation through a gradient boosting model to a small Streamlit app displaying predictions and SHAP explanations.

Model monitoring script

A script that compares data and prediction distributions over time, detects drift and produces a compact report with charts. The team is notified automatically when anomalies appear.

Evaluation harness for a RAG chatbot

A test set-up with a question set, reference answers and automated scoring that compares different retrieval and prompt configurations. Results are summarised per configuration in a clear table.

Prerequisites

Solid programming skills in TypeScript/JavaScript or Python and a basic understanding of HTTP APIs and JSON. Some experience using AI assistants in day-to-day development is helpful. Good Python skills, experience with pandas or Polars and scikit-learn, and a grounding in statistics and machine learning.

Technical requirements

A laptop with Node.js or Python, Git and a code editor, plus access to at least one MCP-capable client such as Claude Code, Claude Desktop, Cursor or VS Code with Copilot.

Data protection & compliance

Training and analysis data often contains personal information, so use schemas, aggregates or synthetic data in prompts and observe GDPR principles such as purpose limitation. For high-risk applications, the EU AI Act sets requirements on data quality, documentation and transparency, and the licences of datasets, pre-trained models and Python packages must also be checked.

Suggested dates

All dates are suggestions – your preferred date is always possible.

  • Tue, 24 November 2026 – Wed, 25 November 2026 Online Request
  • Tue, 1 December 2026 – Wed, 2 December 2026 Hanover Request
  • Wed, 16 December 2026 – Thu, 17 December 2026 Online Request
  • Mon, 11 January 2027 – Tue, 12 January 2027 Online Request
  • Tue, 19 January 2027 – Wed, 20 January 2027 Stuttgart Request
  • Tue, 2 February 2027 – Wed, 3 February 2027 Online Request

Frequently asked questions

Why would I need my own MCP server?

Whenever AI assistants need access to internal systems such as ticketing tools, databases, documentation or line-of-business applications. Your own server defines precisely which actions and data are available and can then be reused across several clients. That avoids isolated one-off integrations and keeps access firmly under your control.

Which programming language does the course use?

You can choose between TypeScript and Python, both of which have official SDKs. The concepts are identical and all exercises are available in both versions. Developers working in other languages still benefit, because the protocol is language-agnostic and SDKs for other languages follow a similar structure.

How secure is MCP for enterprise use?

The protocol provides mechanisms for authorisation and permissions, but real-world security depends on implementation and operations. Risks such as prompt injection through tool results or over-privileged servers are genuine. We cover concrete countermeasures, approval processes for third-party servers and how to handle personal data.

Can I trust AI on statistical questions?

AI assistants know common methods very well but occasionally make methodological mistakes, for example in splitting training and test data or choosing metrics. On the course you learn to ask the AI explicitly about assumptions and risks, check results against baselines and sanity checks, and keep professional accountability firmly with yourself.

Does AI-assisted coding work in Jupyter notebooks?

Yes. Many assistants support notebooks directly, and agents can also read, execute and convert notebooks into modules. We show when notebooks are ideal for exploration and how to make the transition to versioned, tested code efficiently with AI support. The two approaches complement each other.

How much does “Model Context Protocol (MCP) for Data Scientists and ML Engineers” cost?

The course costs a flat €949 plus VAT per participant – whether online or on-site. This includes 2 training day(s) with a live trainer, course materials, a lab environment and a certificate of attendance.

Can I suggest my own date?

Yes. Our suggested dates are just a guide. In the request form you can always describe your preferred date or period in free text – we check availability and confirm within one working day.

Is the course online or on-site?

Both are possible: live online in a virtual classroom or on-site in Munich, Salzburg, Berlin, Augsburg, Hanover, Bremen, Stuttgart or Kaiserslautern. We can also run the course exclusively for your team.

More courses for Data Scientists and ML Engineers

More Model Context Protocol (MCP) courses

All 29 Model Context Protocol (MCP) courses

Model Context Protocol (MCP) · MunichModel Context Protocol (MCP) · SalzburgModel Context Protocol (MCP) · BerlinModel Context Protocol (MCP) · AugsburgModel Context Protocol (MCP) · HanoverModel Context Protocol (MCP) · BremenModel Context Protocol (MCP) · StuttgartModel Context Protocol (MCP) · KaiserslauternModel Context Protocol (MCP) · Online