Senior Data Engineer (Data Platform)

About Gallatin

At Gallatin, we are rebuilding logistics infrastructure for the national security missions of the United States and allied partners. We build AI systems that determine how logistics decisions are made — not just how they're executed. From factory to foxhole, we operate at the layer where data becomes decisions, and decisions make the advantage.

We believe that you won't change the world by phoning it in and as such we work in-office in one of our core engineering locations: El Segundo, CA; Austin, TX; Washington, DC; San Francisco, CA.

About the Role

Logistics decisions are only as good as the data underneath them, and in our world that data is fragmented across legacy systems of record, spreadsheets, slide decks, message traffic, and sensor feeds that were never designed to talk to each other. The Data Platform team connects those sources, cleans and models the data, and serves it to Navigator and our AI agents.

We're hiring a Senior Data Engineer to build that platform. You'll design and ship containerized services that create and manage external data connections, run an open lakehouse, and power the transformations that take raw source data to curated, semantically modeled entities. You'll also build the tools on top: fullstack applications that let logistics SMEs and analysts connect sources, map fields, and test pipeline changes on their own.

The platform runs everywhere Navigator runs: commercial and government cloud, single-tenant high-side enclaves, and air-gapped appliances at forward sites. That means open-source, Kubernetes-native components, hardened images, and data movement that tolerates intermittent and disconnected networks.

Success means a mission partner can point us at a new system of record and have validated, lineage-tracked, query-ready data flowing into Navigator within days.

About You

You start with the catalog: where data came from, what it means, and how much to trust it. You've integrated data from systems that disagree with each other and built the models and mappings that reconcile them.

You're a data engineer with strong software engineering skills, comfortable building a React or Vue front end for the pipeline tools you design. You bias toward action, ship working integrations early and improve them in production, and pull in teammates when the problem crosses into expertise you don't have.

What You'll Do

Build the Data Platform

• Design, build, and operate containerized services that create, configure, and manage external data connections: databases, REST/SOAP APIs, file drops, message buses, and legacy systems of record

• Own our open lakehouse: object storage, Apache Iceberg tables, a REST catalog with data versioning

• Build batch and streaming ingestion with change data capture where it fits, and design it to run on disconnected, degraded, intermittent, and limited (DDIL) networks

• Run the platform on Kubernetes across commercial cloud, government cloud, and air-gapped edge appliances using hardened, FIPS-validated images

Turn Raw Data into Trusted Entities

• Build transformation pipelines that take source data through raw, cleaned, and curated layers into the semantic entity model Navigator and our AI agents run on

• Implement entity resolution and data fusion so each ship, part, unit, or supply request resolves to one record across sources

• Structure unstructured inputs (documents, slides, spreadsheets) into typed, relational records

• Build lineage, provenance, data quality checks, and confidence labeling into every pipeline so downstream answers show their sources and confidence

Build Tools That Put Data in SME Hands

• Build fullstack applications (TypeScript, Vue, Python services) that let SMEs connect sources, map fields, and compose pipelines with low-code workflows

• Give users sandboxes to branch the data, test a pipeline change, compare results, and promote it to production

• Build documented APIs that Navigator features and AI agents consume

Collaborate and Raise the Bar

• Partner with Product, AI/ML, Frontend, and Infrastructure teams on RFCs, data models, and production code

• Work with mission partners and field teams to map how data moves across their systems and translate operational constraints into platform requirements

• Contribute to code reviews, documentation, and CI/CD practices, and mentor engineers as the data team grows

What We're Looking For

Data Engineering Depth

• 8+ years of professional data engineering experience, including hands-on design and operation of production data pipelines and ownership of a data platform or lakehouse

• Hands-on production experience with an open table format (Apache Iceberg preferred; Delta or Hudi fine) and a distributed query engine (Trino, Presto, or Spark SQL), including partitioning, compaction, and schema evolution

• Demonstrated experience with streaming and batch data pipeline technologies (Kafka, NiFi, Debezium, Spark, or similar)

• Strong proficiency in Python and SQL, with experience integrating and validating data from multiple heterogeneous sources (relational and NoSQL databases, APIs, files, and streams)

• Strong data modeling skills for both analytical and operational workloads; experience designing entity models for complex real-world domains

• Production orchestration experience (Airflow, Dagster, Argo Workflows, or similar)

• Production experience writing services in Python and/or TypeScript, plus enough front-end skill (Vue, React, or similar) to ship a usable internal tool end to end

• Comfortable designing and documenting RESTful APIs that other teams build on

Containers and Platform

• Strong Kubernetes and Docker experience: you've packaged, deployed, and operated stateful data services on Kubernetes and can debug networking, storage, and namespace issues

• Comfortable building on open-source tooling for self-hosted and air-gapped deployments

• Experience with AWS data services (e.g., S3, Glue, Aurora) or Azure equivalents and containerized environments (Kubernetes), delivered through DevSecOps/CI-CD pipelines

Communication and Growth Mindset

• Demonstrated ability to work independently against well-defined tasking with minimal oversight, and to communicate effectively with both technical and non-technical stakeholders, including data providers and consumers

• Clear communicator across disciplines, whether writing an RFC, explaining a data model to an SME, or recording a weekly demo of recent work (a Gallatin tradition)

• Always seeking to learn and stay current with industry trends and tools

Bonus Points

• Active SECRET or TOP SECRET clearance

• Data catalog, lineage, and governance experience (Iceberg REST catalogs, Nessie, Polaris, OpenLineage, DataHub, or similar)

• Data versioning and branching (Nessie, lakeFS) for sandboxed pipeline testing

• Apache NiFi or other low-code/no-code pipeline tools, including custom processors

• Entity resolution, semantic data fusion, knowledge graphs, or ontology modeling

• Shipping data workloads into air-gapped or classified environments

• Palantir Foundry or ontology experience

• Data infrastructure for AI: metadata and retrieval that make agents more accurate and explainable

• Operations research, optimization, or constraint-solver background

• Background in defense logistics, sustainment, or supply chain systems

• Contributions to open-source data projects

Why Gallatin?

Join a mission-driven, high-impact, and fast-moving startup where your work directly improves defense logistics readiness for the warfighters of the United States and allied forces.

Work alongside a team of passionate engineers, designers, and industry experts.

Competitive compensation incl. generous options grant, 100% employer-paid health insurance premiums, 401k, opportunities for rapid career growth, unlimited PTO, free lunches and snacks in the office.

This position requires the ability to obtain and maintain relevant security clearances. The successful candidate must be able to work in a classified environment when necessary.

If you're not excited about working hard, and solving the hard problems of building AI decision-support systems that enhance our warfighters, don't apply.

The following ranges are based on the cost of labor across US geographic areas. The base ranges from our lowest geographic market to our highest geographic market.