2 days ago
Sibiu, Romania +2 moreSenior
Responsibilities
- Design and implement enterprise RAG solutions with chunking, metadata, permissions, hybrid retrieval, reranking, grounding, and citations.
- Develop controlled document-generation solutions using templates, citations, and version control.
- Build multi-step AI agents with planning, state, memory, tool calling, routing, structured outputs, retries, and failure recovery.
- Integrate agents with PLM, ERP, CPQ, and case management systems, including controlled write-back through human approval.
- Implement text-to-SQL interactions and connect forecasting, optimization, and document-extraction models as reusable agent tools.
- Evaluate retrieval quality, groundedness, task success, and tool correctness, including detailed failure analysis.
- Implement guardrails, tracing, auditability, monitoring, and observability for LLM and agent actions.
- Collaborate with architects, domain experts, and platform engineers to establish technical baselines and measurable business outcomes.
- Deliver reusable, tested Python code, version-controlled prompts, models, configurations, and technical documentation.
- Support MVP releases, incorporate user feedback, and ensure production handover and monitoring.
Requirements
- At least 10 years of relevant technology experience.
- At least 6 years of experience in data science, machine learning, or natural language processing, including hands-on production work with LLMs.
- At least 4 years of strong hands-on experience with Python and SQL.
- Proven experience delivering production-grade RAG, LLM, or agentic AI solutions.
- Strong knowledge of enterprise RAG, embeddings, vector and hybrid search, reranking, grounding, citations, prompt design, structured generation, tool calling, and agent orchestration.
- Strong understanding of agent state, memory, routing, human approval patterns, failure recovery, evaluation, guardrails, tracing, and observability.
- Experience with cloud deployment, monitoring, and LLMOps practices.
- Strong technical communication, stakeholder influence, ownership, mentoring, and judgment in ambiguous environments.
- Preferred experience with Amazon Bedrock, including Agents, AgentCore, and Knowledge Bases, and with LangChain, LangGraph, CrewAI, or comparable frameworks.
- Preferred experience with Amazon OpenSearch, PostgreSQL with pgvector, AWS S3, Lambda, API Gateway, EKS or ECS, and CloudWatch.
- Knowledge of knowledge graphs, ontologies, process mining, workflow analysis, open-source models, model gateways, MCP, A2A protocols, or document intelligence is advantageous.
- Experience with engineering, energy, regulated, or export-controlled content, consulting, or client-facing delivery is preferred.
- Excellent spoken and written English.
Tech Stack
Categories
About NTT DATA
NTT DATA is a Tokyo‑headquartered, publicly traded IT services firm within the NTT Group. It provides consulting, application development, system integration, cloud and enterprise application services, and managed/outsourcing services for large enterprises and public-sector organizations. The company operates in more than 50 countries and is listed on the Tokyo Stock Exchange.
