Senior Data and AI Engineer (Insurance Domain)

Accord Technologies Inc; Philadelphia (Philadelphia County); On-site; USD 120,000 - 150,000; Full time

Job summary

Accord Technologies Inc is seeking a Senior Data and AI Engineer with expertise in the insurance domain. The role involves owning the full technical stack including architecture, connectors, graph databases, and AI integration. Candidates should have substantial experience with graph databases, semantic web standards, and building RAG pipelines.

This onsite position is located in Philadelphia, PA, requiring immediate availability for NJ and PA based candidates. Strong Python skills and familiarity with Microsoft's data ecosystem are preferred.

Qualifications

• 3+ years of hands-on experience with graph databases in production.
• Experience with semantic web standards and ontologies.
• Demonstrated experience building RAG pipelines.
• Hands-on experience with vector databases like Azure AI Search.
• Ability to operate under accelerated delivery timelines.

Responsibilities

• Own the full technical stack from architecture to UI.
• Collaborate with Data Modeller/Ontologist on implementations.
• Translate conceptual models into technical solutions.

Skills

Graph databases experience

Semantic web standards proficiency

RAG pipeline experience

Embedding and retrieval with vector databases

LLM API integration

Tools

GraphDB

Neo4j

Azure AI Search

Python

Microsoft Fabric

Job description

Senior Data and AI Engineer (Insurance Domain)

Location: Philadelphia, PA

Position type: Onsite role (need NJ, PA based candidates who can join immediately)

Tax type: W2 contract

Candidate should be available to start by next week.

Job Description

The role owns the full technical stack from the architecture slide: connectors and ingestion framework, OneLake Medallion staging, GraphDB triple store, Vector Index, Agentic RAG orchestrator, LLM gateway, guardrails, and the consumption UI with conversational chat, SPARQL trace explainability, and graph explorer.

Knowledge Graph & Semantic Technologies (Must-Have)

• 3+ years hands‑on experience with graph databases (GraphDB, Neo4j, Stardog) in a production or advanced PoC context
• Working proficiency with semantic web standards
• Experience loading, validating, and querying ontologies in a triple store environment
• Familiarity with ontology authoring tools (Prot g , Metaphactory) sufficient to collaborate with the Data Consultant on model iterations

AI / ML Engineering & LLM Integration (Must-Have)

• Demonstrated experience building RAG (Retrieval‑Augmented Generation) pipelines, ideally with agentic orchestration patterns
• Hands‑on experience with vector databases (Azure AI Search, pgvector, Pinecone, Weaviate, or Qdrant) for embedding and retrieval
• Experience integrating LLM APIs (Anthropic Claude, OpenAI GPT, or Azure OpenAI) with prompt engineering, guardrails, and citation enforcement
• Familiarity with NL‑to‑SPARQL or NL‑to‑SQL generation techniques, including few‑shot prompting and schema‑grounding approaches
• Understanding of AI safety guardrails: prompt injection defense, output sandboxing, and confidence scoring

Delivery & Collaboration (Must-Have)

• Comfortable operating in an accelerated 8‑week delivery timeline with weekly milestone gates and hard dependencies
• Ability to work closely with a Data Modeller/Ontologist to translate conceptual models into working technical implementations
• Experience in financial services or insurance data environments is preferred but not required, provided strong technical depth in the above areas

Data Engineering & Microsoft Fabric (Good to‑Have)

• Strong Python engineering skills with experience building data pipelines, ETL/ELT processes, and metadata ingestion frameworks
• Experience with Microsoft Fabric ecosystem: OneLake, Lakehouse, Notebooks, Data Factory / pipelines, and Medallion architecture (Bronze/Silver/Gold)
• Familiarity with JDBC/ODBC connectors, REST API integration, and file parsing (Excel, CSV, JSON) for metadata extraction
• Experience with Trino, Databricks SQL, or equivalent federated query engines