New opportunity

Associate Principal AI Architect (Claude AI)

Unisys · Bangalore, India

About this role

The Associate Principal AI Architect owns technical onboarding and integration of the Anthropic Claude platform and leads knowledge transfer from outgoing providers into the business-as-usual (BAU) operating model. The role is also the architecture-level escalation point for L1.5/L2 support.

Responsibilities

Configure the Claude tenant, API gateway, single sign-on, SOC/SIEM and observability integrations across UAT, staging and production. Administer Anthropic APIs and related Claude products.
Lead a time-boxed knowledge transfer with Accenture and Tribe AI, capturing configurations, the prompt and agent catalog, and integration documentation. Develop and validate support runbooks, contribute to the estate assessment and BAU readiness report, and close documentation gaps.
Provide architecture-level input for Priority 1 and Priority 2 incidents, supporting Response and Restoration Service Levels SL-1 through SL-4. Assist with rollout waves and role-based access control under the customer’s AI governance team.
Track platform features, backlog items and releases for FinOps and adoption reporting. Partner with the Security Architect on usage controls and financial guardrails, participate in the biweekly AI Review Board, and deliver advisory engagements involving prompt-library tuning, solution and agent design, integration guidance, and scoped proofs of concept.

Qualifications

BA/BS degree or an equivalent combination of education and experience; 8–12 years in solution or platform architecture, including AI/ML or GenAI platform integration.
Claude Certified Architect – Foundations certification, or a commitment to obtain it within an agreed timeframe after joining. Experience integrating enterprise SaaS/AI platforms, working with GenAI/LLM platforms, leading knowledge transfer, producing operational documentation, handling complex technical escalations and engaging stakeholders is required.
Preferred: production experience with Claude Code, Claude Cowork, Claude Design or Claude Agents; ServiceNow/ITSM operating models; FinOps for AI/token usage; and moving AI proofs of concept into production.

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

Skills for this role

Anthropic ClaudeGenerative AILarge Language ModelsPrompt EngineeringAI Agent DesignAPI GatewaySingle Sign-OnSIEMObservabilityRole-Based Access ControlEnterprise SaaS IntegrationSolution ArchitectureKnowledge TransferTechnical DocumentationIncident ManagementAI GovernanceFinOpsServiceNowITSMStakeholder EngagementClaude CodeClaude CoworkClaude DesignClaude Agents

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

Platform Transition and Knowledge Transfer
The role requires leading a time-boxed knowledge transfer from outgoing providers (Accenture and Tribe AI) to establish a business-as-usual operating model. Success depends on capturing existing configurations, prompt/agent catalogs, and documentation while identifying and closing gaps in the current estate.
Technical Integration and Administration
You must configure the Anthropic Claude tenant across UAT, staging, and production environments. This includes managing API gateways, SSO, SOC/SIEM, and observability integrations, alongside ongoing administration of Claude products like Claude Code and Claude Agents.
Architecture-Level Incident Support
Serving as the primary escalation point for L1.5/L2 support, you are responsible for providing architectural input for Priority 1 and Priority 2 incidents. This requires maintaining strict adherence to defined Service Level Agreements (SL-1 through SL-4).
Governance and FinOps Alignment
You will partner with the Security Architect to enforce usage controls and financial guardrails. This involves tracking platform innovation and token usage to provide accurate reporting for the AI Review Board and internal FinOps stakeholders.

Likely challenges to think through

Managing Complex Knowledge Transfer
Transitioning from external providers often involves incomplete documentation or undocumented tribal knowledge. Prepare by demonstrating a structured methodology for auditing existing systems and creating robust, validated runbooks that ensure operational continuity.
Balancing Innovation with Governance
As an architect, you must enable rapid AI adoption while adhering to strict security and financial guardrails. Prepare to discuss how you balance technical agility with the requirements of an AI Review Board and enterprise-level RBAC policies.
High-Stakes Incident Escalation
Acting as the L1.5/L2 escalation point for P1/P2 incidents requires deep technical knowledge of the Claude platform and its integrations. Prepare by reviewing your experience in high-pressure troubleshooting and your ability to communicate complex architectural solutions to support teams.

Practice your answers

Technical practice

Enterprise Claude Platform Integration

What to coverDescribe your approach to integrating Anthropic Claude into an existing enterprise environment. Focus on architectural considerations for API gateways, SSO, and SIEM/observability to ensure secure and scalable production deployments.

Knowledge Transfer and Operational Readiness

What to coverHow do you structure a formal knowledge transfer from an outgoing provider to an internal team? Discuss how you would validate L1.5/L2 support runbooks and ensure documentation gaps are closed for BAU readiness.

Incident Management in AI Architectures

What to coverAs an escalation point for P1/P2 incidents, how do you diagnose issues within a GenAI platform? Explain your methodology for troubleshooting latency, model performance, or integration failures while maintaining strict service level agreements.

AI Governance and RBAC Implementation

What to coverExplain how you design and enforce role-based access control (RBAC) for AI platforms. Address how you balance user access needs with the security guardrails required by an AI governance team.

FinOps for AI and Token Usage

What to coverHow do you track and report on AI platform token usage to support FinOps practices? Suggest methods for correlating platform features and release cadences with cost management and adoption metrics.

Prompt Engineering and Agent Design

What to coverDescribe your process for tuning a prompt library and designing AI agents for enterprise use cases. Focus on how you ensure consistency, reliability, and alignment with business requirements during the design phase.

Transitioning AI Proofs of Concept to Production

What to coverWhat are the key architectural challenges when moving an AI proof of concept into a production environment? Discuss how you address scalability, security, and operational support requirements during this transition.

Security and Policy Guardrails

What to coverHow do you partner with security architects to implement usage controls and financial guardrails for LLM platforms? Explain how you translate high-level security policies into technical configurations within the Claude tenant.

Behavioural practice

Managing Stakeholder Expectations in Governance Forums

What to coverDescribe a time you had to present technical AI recommendations to a governance board or non-technical stakeholders. How did you ensure your advisory engagement was understood and accepted?

Leading Through Transition

What to coverProvide an example of a time you led a team through a complex knowledge transfer or platform migration. How did you handle resistance or information gaps to ensure a successful handover?

Handling High-Pressure Technical Escalations

What to coverTell me about a situation where you were the final escalation point for a critical system failure. How did you prioritize your actions to restore service while managing communication with impacted teams?

Balancing Innovation with Operational Stability

What to coverDescribe a situation where you had to balance the need for rapid AI feature adoption with the requirement for stable, secure BAU operations. How did you negotiate these competing priorities?

Sharpen your knowledge

Topics to prepare

Enterprise AI Governance and FinOps
Review the principles of AI guardrails, token-based usage reporting, and financial oversight. Exercise: Draft a framework for monitoring LLM token consumption and setting cost-based alerts to align with the role's requirement for financial guardrails and AI Review Board participation.
ITSM and Support Operating Models
Study the integration of AI platforms into standard IT Service Management (ITSM) workflows, specifically ServiceNow. Exercise: Create a hypothetical L1.5/L2 escalation runbook for a P1 incident involving an Anthropic API failure to demonstrate your ability to manage complex technical support.
Knowledge Transfer and Transition Management
Focus on methodologies for capturing technical configurations and documentation from outgoing vendors. Exercise: Outline a structured 30-day knowledge-transfer plan to audit and validate existing prompt catalogs and integration documentation from a third-party provider.

Skills to revise

Anthropic Claude Platform Administration
Gain proficiency in configuring Claude tenants, API gateways, and SSO. Exercise: Review the official Anthropic documentation for Claude enterprise deployment to prepare for the Claude Certified Architect – Foundations (CCA-F) certification.
AI Agent Design and Prompt Engineering
Practice designing and tuning prompts and agents for enterprise use cases. Exercise: Develop a sample agent design document that includes specific system instructions, integration patterns, and evaluation metrics for a business-process automation scenario.
Observability and Security Integration
Understand how to integrate AI platforms with SOC/SIEM and observability tools. Exercise: Map out a monitoring strategy for an LLM integration that tracks latency, error rates, and unauthorized access attempts to ensure compliance with enterprise security standards.

Your preparation checklist

Review Claude Enterprise Architecture Patterns
Prepare to discuss your experience integrating LLM platforms with enterprise infrastructure, specifically focusing on API gateways, SSO, and SIEM/observability integrations. Be ready to explain how you secure and monitor these deployments in UAT and production.
Prepare a Knowledge Transfer Framework
Outline a structured approach for leading a time-boxed knowledge transfer from an outgoing provider. Define how you would audit existing configurations, validate support runbooks, and identify documentation gaps to ensure business-as-usual readiness.
Develop an Incident Escalation Strategy
Draft a brief explanation of your methodology for acting as an L1.5/L2 architecture-level escalation point. Focus on how you would manage Priority 1 and 2 incidents while maintaining strict adherence to service level agreements.
Formulate AI Governance and FinOps Insights
Prepare to discuss your experience with AI governance, including role-based access control (RBAC) and implementing financial guardrails. Be ready to explain how you track token usage and platform features to report on adoption and costs.
Prepare Thoughtful Questions for the Interviewer
Develop questions regarding the current state of the Claude tenant, the primary challenges identified in the transition from Accenture and Tribe AI, and how the AI Review Board influences architectural decision-making at Unisys.

Availability last checked: 4 Nov 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…