Senior Data Engineer: Agentic Memory & Retrieval

Coforge; Oaks (PA); On-site; USD 130,000 - 190,000; Full time

Job summary

Coforge is seeking a Senior Data Engineer (Agentic Retrieval & Memory) in Oaks, PA for a full-time role. You will design agent memory interfaces across session state, short-term history, and long-term memory, and select optimal stores for each tier.

You will integrate with the enterprise search platform and evaluate backing stores for performance and cost. You will leverage 5+ years in data engineering, 3+ years in agent memory patterns, and 3+ years with vector stores to build scalable,

Qualifications

• 5+ years hands-on data engineering in production environments.
• 3+ years working on agentic or LLM data patterns, with depth in agent memory.
• 3+ years with vector and semantic search using modern vector stores.
• 3+ years designing NoSQL, document, or key-value stores for high-write workloads.
• 2+ years with caching or in-memory stores like Redis.
• 3+ years Python to production standard.
• 2+ years on a major cloud platform.

Responsibilities

• Design and build the agent memory interface across tiers: session state, short-term history, and long-term memory.
• Select and implement appropriate stores for each memory tier (cache, document store, vector store).
• Design and build retrieval interface over the enterprise search platform via shared SDK.
• Assess current backing stores against workload, including partitioning, TTL, retention, read/write patterns, latency, and cost.

Skills

Data engineering

Agent memory patterns

Vector search

NoSQL design

Caching stores

Python production

Cloud platforms

RAG retrieval

Tools

Azure AI Search

PostgreSQL with pgvector

Elasticsearch

Redis

Job description

Coforge is seeking a Senior Data Engineer (Agentic Retrieval & Memory) in Oaks, PA for a full-time role. You will design agent memory interfaces across session state, short-term history, and long-term memory, and select optimal stores for each tier.

You will integrate with the enterprise search platform and evaluate backing stores for performance and cost. You will leverage 5+ years in data engineering, 3+ years in agent memory patterns, and 3+ years with vector stores to build scalable,