Role Summary

We are looking for a hands-on Applied AI Architect to define, design, and guide the delivery of enterprise-grade Generative AI and agentic solutions across our product and solution portfolio. The role sits at the intersection of business problem framing, AI system architecture, engineering leadership, and client engagement.

You will shape how we build multi-agent systems that are reliable, secure, governed, observable, and cost-effective in real enterprise environments. You will work closely with product leaders, Applied AI Engineers, Data Engineers, Cloud Platform Engineers, security teams, and client stakeholders to take solutions from opportunity definition and architecture through production deployment.

This is an architecture role for an experienced practitioner, not a presentation-only role. You should be comfortable producing architectures and decision records, reviewing code, building reference implementations, and guiding teams through difficult technical trade-offs.

Key Responsibilities

- Partner with business, product, and client stakeholders to identify high-value AI use cases, define success measures, and translate business problems into feasible AI solution architectures.
- Design end-to-end architectures for Generative AI applications, agents, and multi-agent systems, covering orchestration, model selection, context, memory, tools, retrieval, identity, governance, evaluation, observability, and deployment.
- Determine when to use deterministic workflows, single agents, or multi-agent systems, and define patterns for supervision, routing, delegation, state, human approval, and failure recovery.
- Architect secure integration of agents with enterprise data, APIs, business applications, MCP servers, and platforms such as Databricks, Snowflake, and cloud AI services.
- Define AI quality and reliability standards, including evaluation frameworks, tracing, regression testing, red teaming, guardrails, monitoring, and continuous improvement.
- Establish responsible AI, security, privacy, and governance controls covering access, data leakage, prompt injection, auditability, model usage, and cost controls.
- Create reference architectures, reusable components, design standards, and architecture decision records for engineering teams.
- Provide hands-on technical leadership through prototyping, code and design reviews, troubleshooting, and coaching Applied AI and engineering teams.
- Support proposals, solution workshops, client presentations, technical due diligence, effort estimation, and delivery planning for AI initiatives.
- Track the evolving model, framework, protocol, and platform landscape, and recommend pragmatic technology choices for products and client solutions.





What We're Looking For

We are looking for proven architecture judgement grounded in hands-on delivery. Candidates should be able to explain not only what they built, but why they chose that design and how it performed.

- Approximately 6-10 years of experience in software engineering, AI/ML engineering, data and AI platforms, or solution architecture, including at least 2-3 years of hands-on Generative AI or agentic system delivery.
- Demonstrated experience architecting and delivering LLM-powered applications or multi-agent systems in production or enterprise pilot environments.
- Solid hands-on Python skills and solid understanding of API design, distributed systems, asynchronous processing, testing, CI/CD, and cloud-native application architecture.
- Deep understanding of LLM concepts, including context engineering, tool/function calling, structured outputs, embeddings, retrieval, model routing, latency, and inference cost.
- Practical experience with agent frameworks and SDKs such as LangGraph, LangChain, CrewAI, Pydantic AI, Google ADK, OpenAI Agents SDK, Claude Agent SDK, Microsoft Agent Framework, or equivalent.
- Strong knowledge of RAG and enterprise grounding, including document processing, hybrid and semantic search, structured data access, re-ranking, citation, and evaluation.
- Experience designing AI evaluation and observability approaches, including evaluation datasets, trace analysis, LLM-as-judge where appropriate, deterministic checks, and regression testing.
- Strong understanding of AI security and governance, including identity, authorization, secrets, data privacy, prompt injection, data exfiltration, auditability, and responsible AI controls.
- Ability to communicate architecture and trade-offs clearly to engineering teams, business leaders, and clients.

Agentic AI Architecture Experience

- Designed a multi-agent system with clearly defined agent responsibilities, coordination patterns, shared context, tool contracts, permissions, and escalation paths.
- Experience designing explicit orchestration, state, memory, checkpoints, retries, human-in-the-loop, and recovery patterns for long-running or multi-step workflows.
- Experience using modern agent development tools and coding agents such as Google ADK, OpenAI Codex,



and Claude Code to accelerate prototyping and engineering.
- Familiarity with MCP, agent-to-agent communication patterns, durable execution, idempotency, timeouts, and asynchronous agent workflows.
- Ability to establish standards for context management, model versioning, tool governance, evaluation gates, release management, and production monitoring.

Nice to Have

- Experience designing and delivering agentic solutions within Databricks or Snowflake.
- Exposure to Databricks Unity Catalog, Model Serving, Vector Search, AI Gateway, MLflow, or Snowflake Cortex Agents, Cortex Analyst, Cortex Search, and Cortex AI Gateway.
- Experience with AWS Bedrock, Azure AI Foundry, Google Vertex AI, or equivalent cloud AI platforms.
- Experience with LangSmith, Langfuse, Arize Phoenix, OpenTelemetry, or similar evaluation and observability tooling.
- Experience in consulting, pre-sales, solution engineering, or client-facing technical advisory roles.
- Relevant cloud, Databricks, Snowflake, or AI architecture certifications.

What Success Looks Like

- You turn ambiguous business opportunities into clear, feasible AI architectures with measurable success criteria.
- Your designs are pragmatic, avoiding unnecessary agent complexity while enabling reliable automation where agents add real value.
- Engineering teams use your reference architectures, standards, and reusable components to deliver AI solutions faster and more consistently.
- Solutions you guide move beyond demos into secure, observable, governed, and supportable production use.

What We Offer

- Opportunity to shape AI and agentic solution architecture across a diverse portfolio of products and client solutions.
- Exposure to multiple industries including Retail & CPG, Media & Entertainment, Sports, Energy, and Telecommunications.
- Hands-on work with frontier foundation models, modern agent frameworks, Databricks, Snowflake, AWS, Azure, and GCP.
- Opportunity to influence AI engineering standards, reusable platforms, and delivery practices for a growing organisation.
- A development path toward Principal AI Architect, Agentic AI Practice Lead, or senior technical leadership roles.

About EXL

EXL (NASDAQ:EXLS) is a global data and AI company that helps clients improve business outcomes through data, analytics, AI, digital solutions, and operations. Our teams work with organisations across industries to design, build, and deploy technology-driven solutions addressing complex business problems. Headquartered in New York, EXL has more than 40,000 professionals in locations throughout the United States, Europe, Asia (primarily India and Philippines), Latin America, Australia and South Africa

📌 Applied AI Architect (Gurugram)
🏢 EXL
📍 Gurugram