16 hours ago
Gurgaon, IndiaMid Level
Responsibilities
- Build, operate, and continuously improve the shared knowledge layer for internal AI assistants.
- Develop ingestion, parsing, chunking, metadata, embedding, and indexing pipelines for unstructured content.
- Create evaluation sets and metrics to measure retrieval and answer quality, diagnose failures, and improve performance.
- Partner with content owners to establish sources of truth, ownership, refresh cycles, and content retirement processes.
- Carry source permissions, sensitivity classifications, and personal-data handling rules through AI retrieval.
- Maintain taxonomies, metadata standards, and entity definitions to classify, filter, and link content to structured records.
- Onboard new AI assistants and use cases onto the shared knowledge layer.
- Monitor ingestion and retrieval pipelines, resolve issues, and escalate where needed.
- Document knowledge sources, ownership, refresh status, access rules, pipeline logic, and evaluation results.
Requirements
- Hands-on experience building RAG or enterprise search pipelines, including PDF, Word, and email parsing, chunking, embeddings, and vector search.
- Experience measuring retrieval and answer quality with test sets and metrics and using results to drive improvements.
- Experience with taxonomies, metadata, content lifecycle management, versioning, and source-of-truth practices.
- Practical understanding of access control, sensitivity classification, and personal-data handling in unstructured content.
- Strong Python and SQL skills.
- Experience with a cloud platform, preferably Google Cloud, and at least one vector store or search engine.
- At least 3 years of experience in data engineering, search or retrieval engineering, or information management.
- At least 1 year of experience building retrieval or document-processing pipelines for LLM applications.
- Degree in Computer Science, Information Systems, Information Science, or a related field.
- Ability to explain technical trade-offs to non-technical content owners and collaborate with engineering teams across locations.
Benefits
- Hybrid working with 2 days working from home and 3 days in the office.
- Staff discounts at Accor hotels worldwide.
- Hands-on work with day-to-day impact across a global hospitality brand.
- Supportive, results-driven team environment.
- Role based in Gurgaon and reporting to the AI & Data Engineering Platform Manager.
Categories
AI ApplicationsData Engineering
About Accor
Accor is a global hospitality group that owns, manages, and franchises hotels across luxury, premium, midscale, and economy segments, alongside the ALL – Accor Live Limitless loyalty program. It operates roughly 5,500 hotels in 110 countries under brands such as Ibis, Novotel, Mercure, Sofitel, Fairmont, and Raffles, generating revenue primarily from management and franchise fees and F&B. Founded in 1967, the company is headquartered near Paris, France.
