New opportunity

ServiceNow Technical Architect — Conversational / GenAI

CGI · Hyderabad, India

About this role

CGI is seeking an experienced ServiceNow Technical Architect – Conversational / GenAI to lead the architecture and implementation of enterprise conversational AI capabilities on the ServiceNow platform. The role involves owning NLU model design, Virtual Agent architecture, conversational workflows, Now Assist/GenAI integration patterns, AI Agent capabilities, and enterprise AI guardrails. The successful candidate will collaborate with engineering, security, privacy, and AI governance teams to ensure solutions are scalable, secure, and auditable.

Key Responsibilities:

Own the architecture and technical design of ServiceNow Virtual Agent and conversational AI solutions.
Design and govern NLU intents, entities, and conversational flows.
Implement integration patterns for ServiceNow Now Assist and Generative AI.
Establish AI guardrails, human-in-the-loop controls, and audit trails for AI-driven actions.
Define KPIs for conversational AI performance and drive continuous improvement.
Mentor development teams and provide architectural guidance.

Required Qualifications

6+ years of overall IT/ServiceNow experience.
Strong hands-on experience with ServiceNow Virtual Agent and NLU implementation.
Practical understanding of Generative AI, LLM integration patterns, and prompt engineering.
Knowledge of enterprise AI governance, responsible AI principles, and risk controls.
Strong solution architecture and stakeholder management skills.

Preferred Qualifications

Experience with ServiceNow AI Agent Studio, AI Agent Orchestrator, and Now Assist.
Knowledge of RAG architectures, vector databases, and AI evaluation frameworks.
Experience with LLM platforms such as Azure OpenAI, OpenAI, Google Gemini, or Amazon Bedrock.
Relevant certifications including ServiceNow Certified System Administrator (CSA), Certified Implementation Specialist – Virtual Agent, or Certified Application Developer (CAD).

For full details about this role, click “Apply on company site” to view the employer’s original posting.

Skills for this role

ServiceNowConversational AIGenerative AILLMPrompt EngineeringAI GovernanceNLUVirtual AgentAI AgentsResponsible AINow AssistIntegrationHubFlow DesignerAzure OpenAIOpenAIGoogle GeminiAmazon BedrockRAGVector DatabasesSemantic Search

Checking your profile…

Preparation suggestions, not the employer’s actual interview questions. Always check the original posting for current requirements.

Understand the role

What matters most

Architecting Secure Conversational AI
The role requires designing and governing Virtual Agent and GenAI solutions that are not only functional but also secure, auditable, and compliant with enterprise AI governance standards. You must demonstrate how to build scalable architecture patterns that integrate human-in-the-loop controls and rigorous AI guardrails.
NLU and Conversational Flow Optimization
You are expected to own the end-to-end design of NLU models, intents, and entities. A core priority is driving continuous improvement in conversational performance metrics such as intent recognition accuracy, containment rates, and user satisfaction by analyzing failures and refining flows.
GenAI Integration and Prompt Engineering
You must define and implement integration patterns for ServiceNow Now Assist and other LLM platforms. This involves establishing best practices for prompt engineering, context management, grounding, and hallucination mitigation to ensure AI-generated responses are reliable and accurate.
Stakeholder and Governance Engagement
As a technical lead, you must bridge the gap between business requirements and technical execution. You will be responsible for representing the conversational AI squad in architecture, security, and risk review forums, requiring strong communication skills to secure approvals for AI use cases.

Likely challenges to think through

Balancing Innovation with Strict AI Governance
The primary challenge is implementing cutting-edge GenAI features while adhering to rigid enterprise security, privacy, and risk-management policies. Prepare by being ready to discuss how you have previously designed 'approval gates' and 'audit trails' for AI-driven actions without stifling the agility of the conversational experience.
Managing Agentic AI Complexity
Moving from simple chatbots to agentic workflows involves complex tool calling and autonomous decision-making. The difficulty lies in defining clear boundaries and permissions for AI agents. Prepare by articulating how you define 'human intervention points' for high-risk tasks to prevent unauthorized or unsafe business actions.
Ensuring Scalability Across Business Functions
Designing conversational solutions that are reusable across diverse business units is difficult. You will likely face challenges in maintaining consistency while allowing for functional flexibility. Prepare to discuss your approach to creating modular, reusable architecture patterns that can be scaled across an entire enterprise.

Practice your answers

Technical practice

Designing ServiceNow Virtual Agent NLU Models

What to coverExplain your approach to designing and governing NLU intents, entities, and utterances. Discuss how you balance intent recognition accuracy with the need for clear, maintainable conversational flows.

Implementing Now Assist and GenAI Integration Patterns

What to coverDescribe how you would architect an integration between ServiceNow and an LLM platform like Azure OpenAI or Amazon Bedrock. Focus on how you manage prompt templates, context, and grounding to ensure relevant, secure outputs.

Establishing AI Guardrails and Risk Controls

What to coverHow do you design guardrails to prevent inappropriate or inaccurate AI-generated responses? Discuss your strategy for implementing human-in-the-loop controls and approval gates for high-risk AI actions.

Architecting Agentic AI Workflows

What to coverDiscuss the architectural considerations for building autonomous AI agents within ServiceNow. Address how you define agent boundaries, tool calling, and the necessary validation points for multi-step task execution.

Measuring Conversational AI Performance

What to coverWhich KPIs do you prioritize to evaluate the success of a conversational AI solution? Explain how you use metrics like containment rate, fallback rate, and user satisfaction to drive continuous improvement.

RAG Architectures and Vector Databases

What to coverExplain the role of Retrieval-Augmented Generation (RAG) in an enterprise ServiceNow environment. Describe how you would leverage vector databases to improve the accuracy and relevance of AI responses.

Ensuring Auditability and Compliance in AI

What to coverHow do you design logging and monitoring systems to ensure AI interactions are fully auditable? Address how you align these designs with enterprise security, privacy, and AI governance standards.

Mitigating GenAI Security Risks

What to coverWhat strategies do you employ to mitigate risks such as prompt injection, data leakage, and hallucinations? Discuss how these security measures are integrated into the overall solution architecture.

Behavioural practice

Navigating Stakeholder Conflicts in AI Governance

What to coverDescribe a situation where you had to balance business requirements for AI speed with strict security or privacy governance constraints. How did you communicate the trade-offs and reach a consensus?

Mentoring Teams on New AI Technologies

What to coverProvide an example of how you have mentored a development team on complex topics like prompt engineering or conversational design. How did you ensure they adopted best practices?

Translating Business Needs into Technical Architecture

What to coverTell us about a time you translated a vague business requirement into a scalable, technical conversational AI solution. What was your process for defining the scope and architecture?

Handling AI Project Failures or Performance Issues

What to coverDescribe a time when a conversational AI solution failed to meet performance targets or user expectations. What steps did you take to analyze the root cause and implement improvements?

Sharpen your knowledge

Topics to prepare

Enterprise AI Governance and Risk Controls
Review frameworks for responsible AI, including human-in-the-loop mechanisms, audit trails, and data privacy compliance. Exercise: Draft a risk assessment template for an AI-driven workflow that includes specific approval gates and fallback logic for high-risk actions, as this is critical for the role's focus on secure, auditable implementations.
Agentic AI and Orchestration Patterns
Study the architecture of autonomous agents, focusing on tool calling, planning, and multi-step task execution within an enterprise ecosystem. Exercise: Map out a conceptual architecture for an AI agent that requires human validation before executing a sensitive ServiceNow transaction, ensuring you can explain the boundary between autonomous and manual steps.
RAG and LLM Integration Architectures
Understand Retrieval-Augmented Generation (RAG) patterns, including grounding, vector databases, and semantic search. Exercise: Design a high-level data flow diagram showing how ServiceNow data is indexed, retrieved, and used to ground an LLM response, specifically focusing on how to mitigate hallucinations and ensure data security.

Skills to revise

ServiceNow Virtual Agent and NLU Design
Refine your ability to design and govern NLU models, intents, and conversational flows. Exercise: Create a complex Virtual Agent topic flow that includes a clear fallback mechanism and an escalation path to a human agent, ensuring the design optimizes for resolution rates and user experience.
Prompt Engineering and Context Management
Develop proficiency in creating prompt templates and managing context for Now Assist or external LLM integrations. Exercise: Write a set of system prompts for a specific business use case that includes constraints to prevent prompt injection and ensures the output remains within defined enterprise guardrails.

Your preparation checklist

Map ServiceNow Virtual Agent Expertise
Prepare a summary of your experience designing NLU models, intents, and entities within ServiceNow Virtual Agent. Be ready to discuss how you have optimized conversation flows, handled fallback mechanisms, and improved resolution rates in previous implementations.
Develop a GenAI Integration Framework
Draft a high-level architectural approach for integrating LLMs (such as Azure OpenAI or Amazon Bedrock) into the ServiceNow ecosystem. Focus on your methodology for prompt engineering, context management, and grounding techniques to mitigate hallucinations.
Define AI Governance and Guardrails
Create a checklist or policy document outlining how you implement human-in-the-loop controls, audit trails, and security guardrails for AI-driven actions. This should demonstrate your ability to balance innovation with enterprise risk, privacy, and compliance requirements.
Prepare an Agentic AI Use Case Walkthrough
Select a past project or hypothetical scenario involving agentic workflows. Prepare to explain your design for tool calling, agent orchestration, and the specific validation points you established to ensure autonomous actions remained within defined authorization boundaries.
Formulate Strategic Questions for the Interview
Prepare questions regarding the current maturity of the team's AI governance framework, the specific business functions currently prioritized for Now Assist adoption, and how the architecture team collaborates with security and privacy stakeholders on high-risk AI deployments.

Availability last checked: 18 Oct 2026. Status can change after a check; confirm on the employer’s website. Report an issue · Our checking policy

KEEP EXPLORING

Similar AI jobs

Related skills and specializations in India. Matched to this role, not your profile.

Explore more jobs

Finding similar opportunities…