We are seeking a hands-on Backend Engineer to build and operate the core AI/ML-backed systems powering BharatIQ’s consumer experiences at scale, requiring prior production delivery of ML-enabled backend platforms (not on-the-job ML learning). The role involves designing and running AI product runtime services including orchestration, session/state management, policy enforcement, and tool integration; implementing end-to-end retrieval and memory pipelines (document ingestion, chunking, embeddings, vector indexing, hybrid search, re-ranking, caching, freshness, and deletion semantics); and productionizing ML workflows with strong online/offline parity, feature and metadata services, model contracts, and evaluation instrumentation. You will own performance, reliability, and cost optimization across latency, throughput, cache efficiency, and infrastructure, with observability-by-default through tracing, structured logging, metrics, guardrails, and resilient fallback paths. The ideal candidate brings 6–10 years of backend engineering experience, including 2–3 years delivering ML/AI-backed products such as search, recommendations, ranking, or RAG systems; strong practical understanding of embeddings, retrieval quality, evaluation metrics, and data drift; deep distributed systems expertise; and the ability to independently design, build, deploy, and operate systems in fast-paced environments. Bonus experience includes agentic runtimes, tool-calling patterns, vector databases (FAISS, Milvus, Pinecone, Elasticsearch), Kubernetes-based platforms, modern observability stacks, and privacy-aware data handling.