Building Enterprise AI Agents with Tencent ADP: RAG, Workflows and Operational Guardrails Training Course

Enterprise AI Agents with Tencent ADP

• The definition of enterprise AI agents and the value they bring
• Tencent ADP features for agent development, knowledge integration, and workflow automation
• Distinguishing agent-based solutions from standard chat applications
• Typical enterprise use cases and key delivery considerations

Designing Agents for Business Processes

• Defining agent roles, boundaries, inputs, and outputs
• Selecting between single-agent and multi-agent architectures
• Structuring prompts, tools, and business rules
• Planning for escalation, human oversight, and system reliability

Creating RAG and Knowledge Workflows

• RAG principles for grounded answers and access to enterprise knowledge
• Preparing documents, policies, and internal content for retrieval
• Designing retrieval flows and response grounding patterns
• Continuously testing and refining answer quality

Orchestrating Workflows and Integrations

• Translating business processes into agent workflows
• Connecting agents to APIs, internal services, and enterprise systems
• Managing decisions, approvals, retries, and fallback mechanisms
• Coordinating handoffs between workflow stages and specialist agents

Implementing Operational Guardrails

• Guardrails for security, privacy, compliance, and policy enforcement
• Mitigating risks associated with unsafe output, prompt injection, and data exposure
• Incorporating approval checkpoints, audit trails, and access controls
• Designing safe response patterns for high-impact business scenarios

Monitoring, Evaluation, and Continuous Improvement

• Tracking quality, latency, costs, and workflow success metrics
• Evaluating agent behaviour across realistic business scenarios
• Diagnosing common RAG, workflow, and orchestration challenges
• Formulating an implementation plan for pilot and production rollout