Data Engineer

Link Logistics•New York, NY

3d•$140,000 - $155,000•Hybrid

About The Position

Link Logistics Real Estate (“Link”) is a leading operator of warehouses and business parks, specializing in last-mile logistics real estate. Established by Blackstone in 2019, the company connects consumption, technology, and the supply chain across its portfolio, which spans half a billion square feet. We leverage our scale, proprietary data and insights, and foundational focus on sustainability to drive success for our customers’ businesses and deliver value for our stakeholders. We put our people, customers, and communities first and find ways to make a conscious, positive impact where we live and work. Every day, we work to reinvent and lead our industry forward by thinking bigger and challenging the status quo. We're hiring a hybrid ML/Backend Engineer to own the intelligence layer of Link's Analytics Engine. This is the connective tissue role: you'll design the pipelines that bring disparate real estate data to life, build the knowledge graphs and retrieval systems that make that data queryable by LLMs, and architect the backend services that surface insights to investment teams in real time. You won't be handed a spec. You'll talk to investment analysts and asset managers, identify where analytical leverage is lost today, and build systems that close that gap — often combining classical ML, graph-based reasoning, and LLM-native workflows in the same solution.

Requirements

4+ years in ML engineering, backend engineering, or a role spanning both
Hands-on experience shipping LLM-powered applications in production — RAG pipelines, prompt engineering, eval frameworks
Strong Python skills; comfortable owning backend services and APIs end-to-end
Experience with knowledge graphs or graph databases (Neo4j or similar)
Proficiency building data pipelines at scale (Spark, Databricks, or equivalent)
Deep sensitivity to data provenance — a track record of building systems that create analyst trust, not just claim it

Responsibilities

Design graph-based data structures encoding relationships across markets, assets, tenants, and transactions.
Build retrieval pipelines (RAG, hybrid search, structured queries) that give LLMs accurate, contextually rich grounding.
Develop rule-based and agentic LLM workflows that automate investment analytical tasks.
Own prompt engineering, eval frameworks, and production reliability.
Build ETL/ELT workflows that ingest, normalize, and enrich large-scale internal and third-party datasets (property records, leasing data, macro signals, alt data).
Every dataset should have a clear owner, update cadence, and quality SLA.
Build systems that make data trustworthy by design: lineage tracking from source to insight, confidence scoring on derived outputs, and clear attribution so analysts always know where a number came from and how fresh it is.
Treat a bad comp or stale signal as a production incident.
Develop and maintain APIs and services that expose ML outputs and structured data to front-end applications.
Prioritize low-latency, reliability, and clean contracts with the application team.
Work directly with investment and asset management teams to understand analytical needs and iterate quickly.
Treat analyst trust as a first-class product requirement.