Thank you for sending your enquiry! One of our team members will contact you shortly.
Thank you for sending your booking! One of our team members will contact you shortly.
Duration 14 hours
Course Outline
- Day 1: Comprehending Claude and its application in daily operations
- Arrival and strategic objectives
- Current operational status of each participant and review of authorized licenses
- Each participant will contribute one specific use case to address over the two-day period
- Capabilities available to Claude at present and identification of its operational limitations
- Exercise: Document the individual use case on a card, specifying the objective, required input materials, and the desired outcome.
- Mechanisms of language models
- Tokens and context windows, defined as the volume of text a model processes simultaneously
- Factors contributing to variable responses to identical inquiries
- Understanding model hallucinations and strategies for mitigation
- Comparison of Haiku, Sonnet, and Opus models based on performance, speed, and usage constraints
- Evaluation of Claude, Copilot, and ChatGPT to determine the most suitable option for specific tasks
- Execute a comparative analysis by resolving the same task using two different models and assessing the results for quality and speed.
- Optimizing instructions and contextual parameters
- Core components of effective prompting: role definition, task description, contextual data, examples, formatting requirements, and constraints
- Enabling Claude to request clarification prior to task execution
- Techniques for summarizing and comparing extensive documents and retrieving specific supporting passages
- Integrating verification and review steps directly into the instructional framework
- Exercise: Refine a suboptimal prompt from daily operations through three iterative rounds and compare the versions for effectiveness.
- Projects as persistent knowledge repositories
- Logical organization of project directives and knowledge assets
- Centralized storage of organizational standards, domain knowledge, and templates
- Segregating collaborative and private workspaces and isolating sensitive information
- Auditing, correcting, and disabling memory functions as necessary
- Exercise: Establish a project structure for the individual use case, incorporating a primary directive and three distinct knowledge files.
- Artifacts, file processing, and data analysis
- Generation of reports, tables, charts, and presentation drafts as tangible artifacts
- Interrogation and comparison of data within Excel, PDF, and Word documents
- Utilizing analytical tools for calculation and implementing spot-checks for accuracy verification
- Exporting, publishing, and distributing artifacts within the team environment
- Exercise: Generate an analytical output, including a chart and a brief commentary, based on an anonymized Excel dataset.
- Outcome: Each participant will have a configured project for their use case, two refined templates, and an initial artifact derived from actual material.
- Arrival and strategic objectives
- Day 2: Expanding Claude capabilities through skills, connectors, Cowork, and automated routines
- Overview of the ecosystem
- Deployment environments: chat interface, desktop application, Claude in Chrome, Cowork, and Claude Code
- Definitions and distinctions between skills, connectors, MCP, and routines
- Conceptual understanding of agents as AI-driven workflows utilizing external tools
- Live demonstration from initial concept to completed skill implementation
- Exercise: Map the individual use case to the appropriate architectural layer and provide justification for the selection.
- Development of the first custom skill
- Structural components: description, applicable scenarios, workflow logic, examples, and templates
- Collaborative development and refinement of skills with Claude
- Testing procedures using three representative real-world cases
- Protocols for sharing and maintaining skills within the team
- Exercise: Each participant will construct a skill tailored to their use case, such as generating responses in organizational style or producing review reports.
- Connectors and integrated services
- Integration of Microsoft 365 or Google Workspace, file storage systems, and calendar services
- Adhering to the principle of least privilege when managing access permissions
- Identifying potential prompt injection attempts in emails and documents
- Exercise: Establish a connector and execute a cross-service task, such as preparing for an upcoming appointment using mailbox and calendar data.
- Claude Cowork and automated routines
- Delegation of tasks to Cowork with access to designated, approved folders
- Monitoring progress, reviewing plans, and evaluating intermediate results
- Scheduling recurring tasks, such as morning summaries or weekly reports
- Implementing approval checkpoints prior to the execution of sending or deletion actions
- Exercise: Configure a routine for daily overviews and collectively evaluate the initial execution output.
- Organizational guidelines and the 30-day action plan
- Establishing policies for permissible data uploads and prohibited content
- Maintaining shared skills and projects and assigning stewardship responsibilities
- Distribution of the model matrix and reference sheet as a handout
- Exercise: Draft a personal 30-day implementation plan containing three specific actionable steps.
- Outcome: Each participant will depart with a functional skill, a connected service integration, a configured routine, and a plan for the subsequent 30 days.
- Overview of the ecosystem
Requirements
- A laptop equipped with a current web browser, a valid Claude account (Pro, Team, or Enterprise tier), the recommended Claude desktop application, and a specific use case drawn from daily operations.
- No programming proficiency is required.
Target Audience
- Personnel in administration, finance, procurement, human resources, project management, and process management who are utilizing or implementing Claude within their organization and seek to leverage capabilities beyond basic chat interactions.
Testimonials (3)
Comprehensiveness about the backround; how to do a prompt.
Olaf Kannt - Helios Klinikum Krefeld
Course - Generative AI and Prompt Engineering in Healthcare
the tips and recommended prompts that we can take away from this training
Lee Mei Lin - ST Engineering IHQ Pte Ltd
Course - InVideo AI: Creating Engaging Short-Form Videos
use of proper and effective prompt
