Neoflow | Calgary or Toronto, hybrid
Pay: CA$100,000 to CA$140,000 per year
Job type: Full-time, Monday to Friday
Office days: Mondays, usually
Data pipelines for actual pipelines
A lot of data work ends in a dashboard. This job ends at the border.
Every day, millions of barrels of crude oil move from Canada into the United States by pipeline, and every shipment raises the same questions. Where did it come from? Who owned it along the way? What’s actually in it? Neoflow answers them with verifiable data. You’ll build the pipelines that make that data trustworthy, turning raw records from shippers, pipeline operators, and refiners into something U.S. Customs and Border Protection (CBP) can rely on.
What Neoflow does
Our platform gives crude oil and natural gas shipments a verifiable digital passport. It records where the product came from, who held it, and what happened along the way, then shares that record with CBP, customs brokers, and refiners in real time.
This isn’t a slide deck. We built it with CBP and the U.S. Department of Homeland Security, and it’s in commercial operation today. Since July 2026, it’s also been part of a live CBP pipeline test under the National Customs Automation Program. We’re SOC 2 Type 2 certified, headquartered in Calgary, and growing in Canada and the United States.
What you’ll work on
A big part of how data moves through Neoflow will be yours, from the moment it lands to the moment someone at the border relies on it.
- You’ll build batch and streaming pipelines for shipment data: delivery tickets, batch measurements, custody and ownership events, and delivery schedules.
- Messy data comes with the territory, so deduplication, entity resolution, schema changes, and quality checks are a big part of the job. (Cubic metres on one side of the border, barrels on the other. One shipper, three spellings.)
- Different parties report on the same shipment, and they don’t always agree. You’ll build the checks that catch quantities that don’t match or custody that skips a step.
- Where LLMs help, you’ll run them in production properly: inputs, orchestration, retries, timeouts, and cost, with a person reviewing the result when the model isn’t confident or the stakes are high.
- Every generated result links back to its source records. Traceability is literally our product, so our own pipelines had better have it too.
- And you’ll keep an eye on all of it in production, with monitoring, evaluation, drift detection, and a safe way to roll back when something goes sideways.
Who you’ll work with
Day to day, you’ll work directly with Patrick Mandic, our CTO, and Mahmoud Alkhraishi, our Director of Engineering. It’s a tight-knit team with no layers in between, so you’ll take work from idea to production and have a real say in what we build next.
About you
Most of this should sound familiar:
- You’ve spent around 5 years building and running data pipelines in production, on Snowflake, BigQuery, Redshift, Databricks, or something like them. You know data modeling, ETL and ELT, and how to make slow things fast.
- Python and SQL are home turf, and TypeScript is a plus if you have it.
- Data quality is part of how you build: tests, contracts, anomaly checks, reconciliation.
- You’ve taken an LLM-powered workflow to production and dealt with what happens when the model gets it wrong.
- Security, privacy, and audit trails matter to you, because our data ends up in front of CBP and has to hold up.
Also nice, not required: AWS, Docker and Kubernetes, Terraform, Airflow or Dagster, Kafka, and vector search or evaluation tooling. Bonus points for experience with energy, logistics, or customs data, or with W3C Verifiable Credentials and EPCIS. Never seen a pipeline delivery ticket? That’s fine. We’ll teach you.
Don’t match every line? Apply anyway. We care more about how you think than a perfect checklist.
What we offer
- Extended health, dental, and vision care
- Life and disability insurance
- Paid time off and a flexible schedule
- Company events and casual dress
Sound like you?
Apply with your resume and a short answer to one question: tell us about a data pipeline you’re proud of, or one that broke and what you changed afterwards. A few sentences is plenty.
If you need an accommodation at any stage of hiring, just ask and we’ll work it out with you.
This posting is for an existing vacancy.
