Senior Software Engineer, Data

Ruxlo•Wall Township, NJ

About The Position

Ruxlo is building a profile page for every one of America's 150M+ properties, creating a durable record attached to the parcel itself. This record includes work history confirmed by owners, contractor details, and neighborhood connections, aiming to be a comprehensive resource for homeowners, contractors, investors, builders, and agents. The company is leveraging AI and data to create a unique category at the intersection of these fields, backed by an experienced SaaS founder. The product is built on a foundation of public records and crowdsourced layers, feeding into a Property Graph that enhances search and AI capabilities. The platform is accessed through an AI-native app, available on web and mobile, with AI-assisted search and a freemium model.

Requirements

Deep, hands-on command of data modeling, including designing normalized 3NF transactional models and denormalized star schemas for analytics.
Fluency with dimensional modeling: facts and dimensions, grain, surrogate keys, conformed dimensions, and slowly-changing-dimension patterns.
Proven experience designing and operating a medallion architecture (bronze/silver/gold or equivalent layered model).
Strong data lakehouse experience, including open table formats (Delta, Iceberg, Hudi, or similar), schema evolution, partitioning, time travel, and storage/compute separation.
A track record of turning messy, inconsistent, multi-source input into a coherent, well-modeled, trustworthy layer.
Experience with transformation frameworks and "analytics/pipelines as code" (dbt or similar), including modular models, tests, documentation, and lineage under version control and code review.
5+ years in data engineering, building and operating production data pipelines and warehouses.
Expert SQL, plus strong coding ability in Python (or a comparable language) for pipelines, transformations, and tooling.
Comfortable owning an existing orchestration and ELT/ETL stack (workflow schedulers, layered pipelines) and raising the bar on it.
Comfortable owning pipeline reliability: idempotency, incremental and backfill loads, retries, and reconciliation.
Data-quality engineering: contracts, tests, freshness/volume checks, and anomaly detection at layer boundaries.
Cost- and performance-aware: partitioning, clustering, file sizing, and query/compute tuning in a lakehouse or cloud-warehouse setting.
Observability and lineage: using production signals to catch drift and regressions early.
Able to own the data platform end to end with limited guidance and establish patterns from scratch on real foundations.
AI-native by default: using AI tools to multiply output and verifying their results.

Nice To Haves

Founding-team or early-stage experience standing up or re-architecting a data platform on a lean team.
Streaming and real-time pipelines, change-data-capture (CDC), and event-driven ingestion.
Proptech, real estate, fintech, or geospatial data experience, especially with large, heterogeneous public and private datasets.
Experience with cloud data platforms and lakehouse engines (e.g., Databricks, Snowflake, BigQuery, Spark, or similar).
Semantic-layer / metrics-layer design and self-serve analytics enablement.
Data governance, cataloging, and PII / access-control patterns for a multi-tenant data product.
Contract testing between producers and consumers, and using production signals to catch model drift early.

Responsibilities

Own the data model, designing and evolving the layered model end to end (raw landing through bronze, silver, and gold) to progressively clean, conform, and make data trustworthy.
Decide where 3NF integrity belongs and where star-schema denormalization better serves analytics.
Run the lakehouse architecture, including open table formats, partitioning, schema evolution, and the storage/compute split, ensuring it remains fast, correct, and cost-efficient.
Build dimensional models, conformed dimensions, and marts to power the app's data-driven features and internal analytics, with clear grain, keys, and slowly-changing-dimension handling.
Own transformation as code, creating modular, tested, version-controlled models with lineage for safe, reviewable, and observable changes.
Define and enforce the data quality bar through tests, contracts, freshness and volume checks, and anomaly detection.
Make modeled data easy and safe to consume through well-documented tables, stable contracts, and appropriate serving patterns.
Partner with integrations and ingestion teams to ensure new sources land cleanly and flow through the layers into the gold models.
Establish modeling conventions, naming, layer contracts, and lakehouse patterns for the rest of engineering to adopt.
Make build-vs-buy calls on the data stack.