Skip to main content
TALENT247 // SPECIALIST AI EXECUTIVE SEARCH & TECHNICAL RECRUITMENT
APP · Applied AI

Senior Multimodal RAG & Search Systems Engineer (Contract)

Dublin, Ireland · Contract · Hybrid

AI Specialism

Applied AI

Workplace

Hybrid

Location

Dublin, Ireland

Compensation

€95k – €125k

annual package

Role Dossier & Specification

REF: SENIOR-MULTIMODAL-

THE CLIENT & MANDATE

Our client is an acclaimed European enterprise AI lab building next-generation multimodal knowledge platforms and autonomous workplace agents for global organizations. With R&D roots in Northern Europe and a growing engineering hub in Dublin, their platform connects live across enterprise document repositories, spreadsheets, slide decks, codebases, and recorded meetings so teams can delegate complex analytical and operational workflows to trustworthy agents.

They have partnered with Talent247 to bring on a Senior Multimodal RAG & Search Systems Engineer in Dublin on an initial 6-month contract (with a strong pathway to permanent conversion or extension). You will join their Retrieval & Knowledge Infrastructure squad to scale real-time, permission-mirrored hybrid search across tens of millions of heterogeneous enterprise artifacts.

KEY RESPONSIBILITIES

  • Architect and scale low-latency hybrid retrieval pipelines combining lexical BM25, dense embeddings, and late-interaction/cross-encoder reranking across text, tables, slide visuals, and meeting transcripts.
  • Build ironclad, zero-leak Access Control List (ACL) security-trimming and permission-graph synchronization across Google Workspace, Microsoft SharePoint, Slack, Jira, and Salesforce connectors.
  • Design incremental ingestion, semantic chunking, layout-aware parsing, and deduplication pipelines capable of reflecting upstream enterprise document edits and permission revocations within seconds.
  • Implement online retrieval telemetry, click/citation feedback loops, and automated LLM-judge evaluation suites to systematically eliminate retrieval blind spots on complex multi-hop queries.
  • Optimize vector database shard placement, quantization settings, caching tiers, and query planning to hold p95 retrieval latency below 80ms at multi-terabyte tenant scale.
  • Collaborate directly with agent-orchestration engineers to expose rich search and filter primitives to autonomous planning loops.

TECHNICAL ENVIRONMENT & OFFERING

Their stack features Python, Rust, TypeScript, Vespa/Qdrant/Elasticsearch, Kubernetes on GCP, Temporal, and PyTorch embedding services. Operating hybrid in central Dublin, this contract role pays an annualized equivalent of €95,000–€125,000 EUR (structured as competitive daily/contract compensation) with high likelihood of long-term extension or permanent conversion.

Core Requirements

ESSENTIAL QUALIFICATIONS

  • 5+ years of backend software and search/ML systems engineering experience building production search, RAG, or distributed data indexing platforms.
  • Deep hands-on expertise with modern vector and hybrid search engines (Vespa, Qdrant, Weaviate, OpenSearch/Elasticsearch, or pgvector at scale).
  • Strong programming skills in Python and either Rust, Go, or TypeScript for high-concurrency ingestion and query microservices.
  • Direct production experience implementing document-level ACL / RBAC permission filtering in enterprise search or SaaS connector pipelines.
  • Solid grasp of chunking strategies, multimodal PDF/slide parsing, reranking tradeoffs, and quantitative IR evaluation metrics (NDCG, Recall@K, MRR).

PREFERRED / BONUS EXPERIENCE

  • Experience with late-interaction retrieval (ColBERT/ColPali-style visual document retrieval) or custom query-understanding SLMs.
  • Familiarity with Temporal workflows, Kafka/PubSub streaming ingestion, and GCP Kubernetes deployments.

Position Telemetry

Location Dublin, Ireland (Hybrid)
Department Applied AI
Employment type Contract
Workplace type Hybrid
Salary range
€95k – €125k annual package