Software Engineer (Spatial Data Ingestion & Conflation Engine)

Augmodo is building the 'operating system for the physical shelf.' We use spatial computing and wearable AI to provide real-time, store-level insights that were previously impossible to capture. We are moving fast, and our data needs to move even faster—while remaining hyper-accurate.

The Challenge

We are looking for a Software Engineer to own the logic behind our Spatial Data Ingestion & Conflation Engine. You will build the pipelines that take raw, complex data from the edge and transform it into high-fidelity insights for our customers. This role is about precision at scale: ensuring that when a brand or retailer looks at our data, it is versioned, reliable, and 'production-ready.'

Key Responsibilities

• Build Staged Pipelines: Design and maintain versioned data pipelines that handle the ingestion and conflation of complex retail data (e.g., matching computer vision detections to master product catalogs).

• High-Fidelity Output: Ensure that data outputs surfaced to consumer-facing dashboards meet a high quality bar for accuracy and reliability.

• Hybrid AI Strategy: Implement a 'right tool for the job' approach—building robust rules-based normalization and RegEx logic for bounded tasks, while strategically integrating LLMs to solve high-complexity data matching and entity resolution.

• Scale & Cost Management: Architect solutions that are mindful of the massive scale of retail environments, ensuring AI integrations are cost-effective and performant.

• Versioned Data Evolution: Manage the lifecycle of data schema and pipeline logic to allow for rapid iteration without breaking downstream consumer insights.

Required Qualifications

• The 'Data Plumber' Mindset: Extensive experience building reliable, staged pipelines for complex ingestion tasks.

• Entity Resolution & Conflation: Experience merging disparate, messy data sources into a unified 'Golden Record.'

• Pragmatic AI Experience: Comfortable with LLMs and prompt engineering but skeptical enough to know when a simple regex or bounded heuristic is the better engineering choice.

• Engineering Rigor: Belief in versioning everything—from code to data schemas.

Bonus Points

• Physical Product Domain: Experience in retail, logistics, or supply chain (handling SKUs, UPCs, and physical inventory data).

• Spatial Data: Experience working with data that has a physical or geographic component (GIS, LIDAR, or Computer Vision metadata).