AI Architect

About the role

Zensar Technologies is hiring an AI Architect in Pune, Maharashtra, India to shape and evolve the ZenseAI.QI and ZenseAI.AssureAI platform suite. The role designs archetype-specific assurance lifecycles for Classical ML, Generative AI, and Agentic AI, owns evaluation-driven development and release gating, and leads presales, client proposals, and executive-facing value articulation. The architect also builds practice IP, mentors emerging AI-specialist roles, and represents Zensar in client captives and industry forums.

What you’ll do

• Shape ZenseAI.QI and ZenseAI.AssureAI and evolve the platform suite across both engines, including the 18-agent ZenseAI.QI stack and the four-pillar ZenseAI.AssureAI framework
• Design and refine archetype-specific assurance lifecycles for Classical ML, Generative AI, and Agentic AI, including the eight-axis agentic trajectory scorecard
• Own technical roadmap decisions for accelerators such as the Agentic Foundry and the curated, swap-ready tooling ecosystem around each archetype
• Build and govern eval suites that gate every release, and own trajectory grading, red/purple/blue-team probes, safety attestations, drift monitoring, and fairness audits
• Translate evaluation results into release decisions — eval-threshold gates, red-team severity floors, canary and shadow deployments, and rollback rehearsals
• Bring evaluation-driven development practice into client engagements, showing how a live harness beats a one-time audit
• Respond to RFPs, RFIs, and client proposals across both engines, and architect engagements across AI QA Assessment, AI QA Transformation, and Managed AI QA
• Build and deliver executive trust scorecards and portfolio risk heat maps, and represent Zensar in client captives, sales pursuits, and industry forums

What you’ll bring

• Approximately 17–21 years of overall experience, with substantial recent depth in AI/ML systems, LLMs, and agentic architectures
• Deep expertise across Classical ML, Generative AI, and Agentic AI assurance and evaluation
• Ability to keep platforms LLM-agnostic and deployable on any client stack — on-prem, cloud, or hybrid
• Experience designing and governing eval suites, including ground-truth Q&A sets, LLM-as-judge rubrics, and frozen baselines
• Ownership of trajectory grading, red/purple/blue-team probes, safety attestations, drift monitoring, and fairness audits
• Experience translating evaluation results into release decisions such as eval-threshold gates, canary/shadow deployments, and rollback rehearsals
• Strong presales and proposal skills — responding to RFPs/RFIs and translating client requirements into defensible solution architectures and commercial structures
• Ability to present value propositions directly to senior client stakeholders, including CTOs, Chief Risk Officers, and Chief AI Officers

Skills