Context : We need solid Java engineers to build high-throughput, low-latency backend services for AI-powered search query understanding and customer personalisation. These services integrate LLMs for query classification and expansion, and connect with personalisation systems to make search results increasingly relevant to individual customers. The services must operate at very high scale with strict latency requirements, and must be LLM-provider agnostic.Essential Skills : - Java (17+) : Production experience building high-throughput, low-latency services in Java on Spring Boot or equivalent.- LLM Integration : Hands-on experience integrating large language models (GPT-4, Claude, Gemini, open-source models) into backend services - including prompt management, structured output parsing, retry/fallback logic, and token-cost optimisation.- Prompt Engineering : Demonstrated ability to design, version, and evaluate prompts for query understanding, query expansion, and classification tasks. Experience with few-shot, chain-of-thought, and constrained-output prompting patterns.- NLP / Information Retrieval : Strong grounding in query understanding, query rewriting/expansion, synonym handling, intent classification, and search relevance.- Microservices Architecture : Production experience designing stateless, cache-first, horizontally scalable microservices.- Caching Strategies : Experience with distributed caching (Redis, Memcached, or similar).- API Design (RESTful / gRPC) : Clean API contract design, versioning, backward compatibility.- Personalisation Systems :



Demonstrable production experience building or integrating with personalisation engines, recommendation systems, or customer profile/preference platforms.- Data Pipeline Engineering : Experience building real-time and batch data pipelines (Kafka, Event Hubs, or similar).- LLM/Agent Memory & Context : Understanding of how to integrate user profile/preference data into LLM agent workflows.- Customer Data Platforms : Experience with customer data platforms, identity resolution, and privacy-compliant data handling (GDPR).- Cloud Platform (Azure) : Production deployment experience on Azure or equivalent.- CI/CD & Observability : Experience with production CI/CD pipelines, structured logging, distributed tracing (OpenTelemetry or equivalent), and monitoring/alerting.Desirable Skills : - E-commerce Search : Prior experience with product search, catalogue search, or retail search systems.- Elasticsearch / OpenSearch : Understanding of search engine internals, index design, and query DSL.- A/B Testing Frameworks : Experience instrumenting and analysing A/B tests.- Model Serving : Experience with model serving infrastructure (TensorFlow Serving, TorchServe, vLLM).- NLP for Preference Extraction : Experience using NLP techniques to derive preferences from unstructured text.- Feature Stores : Experience with feature stores for serving personalisation features at low latency.- Privacy Engineering : Experience implementing data minimisation, consent management, and right-to-erasure.- Grocery Retail Domain : Familiarity with grocery retail personalisation. (ref:hirist.tech)

📌 Publicis Sapient - Manager - AI Java Technology (Bengaluru)
🏢 Publicis Sapient
📍 Bengaluru