Sr. AI Architect

Senior AI Architect

Experience: 9–10 years (with 3+ years architecting and shipping production AI systems)

Location: Chennai, India

Role Summary:

You will own the technical architecture of our enterprise AI platform. The core of the work is agentic AI and Generative AI: designing multi-agent systems, retrieval pipelines, and LLM-backed services that run reliably in production. You will set technical direction, make build-vs-buy calls, and hold the bar on scalability, latency, cost, and quality. This is a hands-on architecture role, not a pure people-management role.

Key Responsibilities:

• Architecture & design. Own the end-to-end design of AI/ML solutions across business domains. Define system boundaries, data flow, agent orchestration patterns, and integration contracts. Document and defend the trade-offs behind each decision.
• Agentic & Generative AI. Architect multi-agent systems and RAG pipelines. Decide where agents, tools, and deterministic code each belong. Design LLM-backed services (chatbots, summarization, text-to-SQL, document intelligence) for accuracy and predictable behavior.
• MLOps & lifecycle. Establish CI/CD for models and agents, versioning, evaluation, monitoring, and retraining strategy. Treat evaluation and observability as first-class, not afterthoughts.
• Cloud deployment. Deploy and optimize on Azure (Azure OpenAI, AI Search, Fabric/Synapse, Data Factory) or AWS/GCP. Own the cost and latency profile of what ships.
• Data & retrieval. Design ingestion and feature pipelines over structured and unstructured data. Architect vector storage and retrieval (Azure AI Search, FAISS, Pinecone, ChromaDB) for quality and scale.
• Performance & cost. Optimize for low latency, high throughput, and unit economics that survive scale-up.
• Stakeholder translation. Work with product, business, and engineering to turn ambiguous problems into feasible, scoped AI solutions. Say no to AI where it does not earn its place.
• Technical leadership. Set standards, run design and code reviews, and mentor senior and mid-level engineers. Raise the team's ceiling, not just your own output.

Required Skills:

Architecture & systems:

• Proven track record designing and shipping production AI systems that real users depend on — not just notebooks or proofs of concept.
• Strong system design: API design, data modeling, concurrency, caching, and failure handling.
• Experience designing agentic systems (orchestration, tool use, state, evaluation) and RAG architectures.

Generative AI & ML:

• Deep, applied knowledge of LLMs: prompting, orchestration, retrieval, fine-tuning, and evaluation.
• Strong grasp of NLP (Transformers, BERT, GPT-family, embeddings) and core ML (supervised/unsupervised methods, ensemble models like XGBoost/LightGBM).
• Working knowledge of a deep learning framework (PyTorch or TensorFlow).

Engineering stack:

• Expert Python, including production-grade API development (FastAPI or Flask) and async/concurrency.
• Agentic frameworks: LangChain/LangGraph, CrewAI, LlamaIndex, or equivalent. Familiarity with Model Context Protocol (MCP).
• Data: SQL (complex query authoring), NoSQL, and vector databases.
• MLOps: Docker, CI/CD, MLflow/Kubeflow, Kubernetes.
• Git-based workflows and Agile/Scrum delivery.

Cloud:

• Production experience on at least one major cloud (Azure preferred). Hands-on with managed AI services and, ideally, Fabric Lakehouse / data platform tooling.

Preferred Qualifications:

• Distributed data processing (Spark, Databricks).
• Open-source contributions or AI/ML publications.
• Computer Vision (CNNs, object detection, ViT, OpenCV). Relevant only if the role's workload includes vision; do not screen on this otherwise

please share CV to [email protected]