Role Objective

We are seeking a knowledgeable and innovative Cloud Data & AI Architect to join our dynamic team at LUMIQ. Data is the core of what we do. The ideal candidate will design and implement large-scale cloud data platforms for our BFSI clients — Ingestion and CDC, Lakehouse and Open Table formats, Transformation, Governance and serving — and then extend that foundation into production agentic AI systems that run underwriting, claims, onboarding and servicing workflows.

You will collaborate with cross-functional teams, assess the technological landscape, and create architecture that aligns with our business objectives and client needs.

Key Responsibilities:

Cloud Solution Design and Development:

- Design, develop, and implement scalable, secure, and flexible cloud architectures.
- Translate business requirements into cloud solutions, ensuring alignment with organizational objectives.
- Lead the deployment of applications, data platforms and AI workloads in cloud environments.
- Good knowledge of at least two of (w.r.t data projects) – Azure, AWS, GCP, Snowflake, Databricks

Data Platform Architecture and Engineering:

- Architect ingestion at enterprise scale — batch, change data capture and near-real-time streaming from core banking, policy administration, claims and third-party systems.
- Design the lakehouse — open table formats, layered storage (raw, cleansed, curated, consumption), schema evolution, partitioning and compaction.
- Own transformation and orchestration design — modelled, tested and version-controlled pipelines rather than hand-written one-offs.
- Choose storage, compute and serving options against latency, concurrency and cost, and keep optimizing — cost is a standing client conversation.
- Data governance, cataloguing, lineage, data quality and observability, with fine-grained access control, PII handling and masking.
- Build reusable industry data models and accelerators that shorten every subsequent implementation.

Agentic AI and GenAI Architecture:

- Architect production agentic systems with the orchestration layer decoupled from the inference layer,



deployed on containers inside the customer's private network.
- Design the data and context layer agents depend on: retrieval over governed sources, feature and context engineering, and document AI / IDP pipelines.
- Define the boundary between agent autonomy and deterministic control — scoped tools, confirmation gates, idempotency and human-in-the-loop.
- Handle long-running workflows properly — durable agent state, checkpointing, resumability and message-history management across multi-step runs.
- Build explainability, auditability and evaluation in from day one: immutable request and response trails, model versioning and rollback, groundedness and drift measurement.

Pre-sales Support:

- Valuable at deriving BOM (Bill of Material) and effort estimate with minimal information, for both data and agentic workloads.
- Conduct and lead POCs, customer demos, and architecture and security reviews.
- Partner with cloud and platform vendors on co-sell, funding programmes and joint solution reviews.

Technology Evaluation and Adoption:

- Evaluate emerging cloud, data and AI technologies, services, and solutions.
- Recommend adoption of new technologies and best practices, and build reusable accelerators that shorten delivery.
- Stay informed on industry trends in cloud, data engineering, agentic AI and the BFSI sector.

Collaboration and Stakeholder Engagement:

- Work closely with data engineers, data scientists, AI engineers, analysts and client IT teams to understand their needs and challenges.
- Communicate complex data and AI solutions effectively to both technical and non-technical audiences.
- Engage in cross-functional projects and initiatives to ensure architecture supports and advances business objectives.

Security and Compliance:





- Implement and maintain cloud, data and AI security best practices to safeguard sensitive data.
- Ensure compliance with BFSI regulations and standards — data residency, RBI/IRDAI guidelines, PII handling and auditability.
- Collaborate with security teams on regular assessments, and on AI-specific risks such as prompt injection, malicious documents, data leakage and model governance.

Performance Optimization:

- Monitor and optimize data pipelines, query workloads and inference infrastructure for performance, cost, and scalability.
- Troubleshoot and resolve issues across cloud environments, data pipelines and agent runtimes.
- Provide technical guidance and support to development teams.

Skills & Qualification

- Bachelor's or master's degree in computer science, Information Technology, or a related field.
- Minimum of 8 years in cloud architecture, design and deployment, with real depth in large-scale data platforms and recent hands-on exposure to AI/GenAI workloads.
- Strong knowledge of major cloud platforms and their managed data and AI services.
- Strong SQL and Python, and infrastructure as code (e.g., Terraform).
- Hands-on with distributed processing, modern lakehouse and open table formats, and workflow orchestration.
- Working knowledge of containerization and orchestration (e.g., Docker, Kubernetes).
- Familiarity with LLM application patterns — retrieval, tool / function calling, agent orchestration, durable state, and evaluation.
- Understanding of BFSI industry compliance, regulations, and security requirements.
- Excellent communication, collaboration, and problem-solving skills.

Nice to Have:

- Cloud architect certification from a major provider, plus a data engineering, machine learning or GenAI speciality.
- Experience with DevOps practices and CI/CD pipelines.
- Experience taking a data platform or an agentic system into production in a regulated environment.
- Open-source contributions to data or agent tooling.
- Knowledge of BI and semantic-layer tooling, vector stores and LLM observability.

📌 Data Architect (Noida)
🏢 Lumiq
📍 Noida