AI Engineer: Multi-Model Agents & Document AI

Xebia DACH; United States; Remote; USD 120,000 - 210,000; Full time

Job summary

Xebia is seeking an experienced AI/agent engineer to join our team building advanced AI and agent-based solutions for financial services domains. You will design and implement LLM workflows, integrate models, and build permissioned tool interfaces to ensure secure, auditable operations.

You will work with Document AI capabilities (OCR, PII masking), retrieval architectures using embeddings, and evaluation frameworks with observability.

Qualifications

• Hands-on experience building AI or agent-based solutions.
• Experience designing or implementing LLM/AI agent workflows.
• Experience with model integration and routing considering cost, latency and risk.
• Experience designing controlled tool interfaces for AI agents.
• Knowledge of MCP-based tool interfaces or permissioned agent-tool integration.
• Experience with Document AI capabilities such as OCR, structured extraction, PII detection or masking.
• Experience with retrieval architectures using embeddings and vector or lexical search.
• Experience designing evaluation approaches for AI systems.
• Understanding observability requirements for agent behaviour.
• Ability to design AI solutions using permissioned, logged and auditable interfaces.
• Ability to collaborate across prototype and production-oriented engineering streams.

Responsibilities

• Develop agents for credit, KYC/compliance, document handling and exception triage.
• Implement multi-role decision patterns for high-impact cases (analyst, critic, guardian).
• Develop a model gateway routing requests between models by cost, risk and latency.
• Support self-hosted models where data residency requires it.
• Build MCP-based tool interfaces restricting actions to permissioned, logged tools.
• Develop Document AI capabilities: OCR, information extraction, PII detection and masking.
• Build retrieval and context using embeddings, vector/lexical search and a knowledge graph.
• Create evaluation framework for AI behavior: evals, A/B testing of models and prompts, and observability.
• Collaborate with Foundation and Lab streams to implement AI concepts on the platform.

Skills

AI agent development

LLM workflows

Model routing

Routing costs/latency

Permissioned tool interfaces

Document AI (OCR/PII)

Embeddings retrieval

AI evaluation

Observability

Auditable interfaces

Cross-team collaboration

Job description

Xebia is seeking an experienced AI/agent engineer to join our team building advanced AI and agent-based solutions for financial services domains. You will design and implement LLM workflows, integrate models, and build permissioned tool interfaces to ensure secure, auditable operations.

You will work with Document AI capabilities (OCR, PII masking), retrieval architectures using embeddings, and evaluation frameworks with observability.