Solution Architect-Data And AI

ZF Group; Hyderabad; On-site; INR 4,500,000 - 6,000,000; Full time

Job summary

ZF India Pvt. Ltd. Hyderabad seeks an Architect – Data & AI to own end-to-end architecture spanning Azure cloud, data platforms, ML/LLM systems, and agentic workflows from concept to deployment. You will shape robust RAG pipelines, vector stores, and production-grade tooling for scalable AI applications.

You will lead architectural discussions with product and engineering leadership, ensure security and cost discipline, and mentor teams while delivering reliable AI platforms in production.

Qualifications

• 11–12 years building Data & AI systems with production deployments of ML and agent-based AI.
• Strong hands-on Python and data tooling with Azure cloud native architectures.
• Expertise in agent design, context engineering, retrieval strategies, and memory management.
• Experience with data mesh or domain data platforms and production security/compliance.

Responsibilities

• Own architecture and design of production-grade Data & AI platforms across cloud, data, ML/LLM systems.
• Architect AI agents, context pipelines, and robust retrieval/high-availability workflows.
• Lead data platform architecture with data mesh, Delta Lake, Databricks, Spark, and governance.
• Drive production deployments, monitoring, cost controls, and stakeholder-aligned trade-offs.

Skills

Python

Databricks

Spark

Delta Lake

Azure

Agent design patterns

Context engineering

Retrieval strategies

Memory management

Production deployment

Tools

LangGraph

LangChain LCEL

Semantic Kernel

Azure ML

MLflow

AKS

Azure Data Factory

Pinecone

Weaviate

pgvector

CI/CD

Job description

Job Description:

Req ID 83710|Hyderabad, India,ZF India Pvt. Ltd.

Job Description

What you can look forward to as an Architect – Data & AI:

• End to End Architecture Ownership: Own architecture and design of production grade Data & AI platforms spanning cloud infrastructure (Azure), data platforms, ML/LLM systems, and agentic workflows—from concept through deployment, operation, and evolution.
• Agentic AI Systems (Production First): Architect and drive AI agents using LangGraph, LangChain Expression Language (LCEL), Semantic Kernel, or similar frameworks, emphasizing context engineering, tool routing, memory strategies, guardrails, and deterministic workflows over prompt hacks.
• Context & Retrieval Engineering: Design robust RAG and non RAG context pipelines using vector databases (Azure AI Search, Pinecone, Weaviate, pgvector), embeddings lifecycle, chunking strategies, re ranking, caching, and fallback logic that work at scale.
• Model & Agent Evaluation: Define and operationalize agent evaluation frameworks (offline + online), including regression tests, task success metrics, hallucination detection, latency/cost controls, and MCP style orchestration patterns for controlled agent execution.
• Data Platform Architecture: Architect data mesh aligned platforms with domain data products, Delta Lake, Databricks, Spark, event driven pipelines, relational and NoSQL databases, and clear ownership, governance, and interoperability models.
• Production Deployment & Ops: Lead production deployments using Azure ML, Databricks, AKS, MLflow, CI/CD, feature stores, monitoring, logging, and incident ready observability for both ML models and AI agents.
• Technical Leadership & Stakeholder Management: Act as a senior architect across teams—review designs, mentor engineers, align with product and business stakeholders, and make explicit trade offs across cost, risk, performance, and maintainability.

Your profile as an Architect – Data & AI :

• 11–12 years building Data & AI systems, with proven production deployments of ML and agent based AI systems (not PoCs or demos).
• Strong hands on experience with Python, data tooling (Databricks, Spark, Delta Lake, Azure Data Factory), and cloud native architectures on Azure.
• Demonstrated expertise in agent design patterns, context engineering, retrieval strategies, tool calling, memory management, and failure mode handling.
• Solid grounding in data architecture and databases (SQL, NoSQL, lakehouse, vector stores), including experience in data mesh or domain oriented data platforms.
• Experience operating AI systems in real environments—security, access control, cost management, latency, reliability, and compliance are second nature.
• Comfortable leading architecture forums, influencing technical direction, and working closely with product, business, and engineering leadership.
• Pragmatic mindset: focused on what runs reliably in production, not chasing the latest Generative AI buzzwords.

Contact

Rajesh Geddam