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: Interactive Chat, Intelligent Assistants, and Knowledge Management
- Langdock Overview
- Platform interface, navigation structure, and configuration settings
- Access to models from Anthropic, OpenAI, Google, and Mistral within a unified environment
- Rationale for adopting a centralized tool instead of individual licenses
- Exercise: Address an operational task using two models and provide a justification for the preferred selection.
- Productive Utilization of Chat
- Generation of texts, summaries, and tabular data
- Processing files directly within the chat interface
- Development of prompt templates for repetitive tasks
- Exercise: Develop a custom template for a standard task and validate it against two scenarios.
- Knowledge Folders and Project Management
- Folder architecture, source integration, and supported document types
- Generating responses based on proprietary documents with cited sources
- Understanding limitations, including context gaps and potential hallucinations
- Exercise: Create a knowledge folder containing three documents and formulate ten test queries.
- Assistant Development
- Configuration of instructions, knowledge bases, model selection, and conversation starters
- Sharing assistants within teams, managing permissions, and controlling visibility
- Application examples such as quotation generation, report templating, and internal FAQs
- Exercise: Build and share an assistant tailored to a common request within the user’s functional area.
- Deliverable: Each participant will possess a prompt template, a knowledge folder, and a shared assistant for their specific operational area.
- Langdock Overview
- Day 2: Integrations, Workflow Automation, and Operational Governance
- System Integrations
- Utilizing SharePoint, OneDrive, and Teams as data sources
- Additional integrations including Confluence, Google Drive, or CRM systems
- Verification of access rights and permission settings
- Exercise: Connect an authorized file storage system and enable assistant access to it.
- Workflow Automation
- Transitioning from basic chat to repeatable automated workflows
- Structuring workflows with triggers, logic, actions, and controls
- Implementing human approval gates, termination criteria, and retry mechanisms
- Exercise: Construct a workflow that categorizes incoming requests and generates response drafts.
- Data Protection and Governance Frameworks
- Data storage protocols, encryption standards, and prohibition on model training usage
- Defining permissible and impermissible data uploads
- Establishing roles, workspaces, and approval processes
- Exercise: Draft a one-page usage policy for the participant’s operational area.
- Tool Selection and Deployment Strategy
- Comparative analysis of Langdock, Claude, and Copilot
- Progression from pilot programs to team adoption and standardization
- Identification of top use cases and development of a 30 to 60-day implementation plan
- Exercise: Prioritize three use cases for the operational area and define subsequent action steps.
- Deliverable: Each participant will have a functional workflow with approval controls, a connected knowledge source, and a 30 to 60-day strategic plan.
- System Integrations
Requirements
- A Langdock account or test environment, a laptop, and one or two anonymized sample documents from daily operations.
- No programming experience is required.
Target Audience
- Departments and key personnel in organizations that utilize or are implementing Langdock as a central AI platform for government and public sector operations.
Testimonials (1)
the tips and recommended prompts that we can take away from this training
