Data Engineer - Data Platform

About the role

Teza's Data Platform team owns the data the firm trades on: every backtest, every live strategy, every portfolio decision starts with data we ingested, cleaned, stored, and served.

The scale, in plain numbers:

• 1 PB of raw historical vendor data, growing by ~150 GB every day

• 120 TB of processed, query-ready data in historical storage

• Thousands of scheduled jobs: run by cron today, actively migrating to Apache Airflow

• Alternative data delivered directly into the real-time feeds of live trading strategies

This is a hands-on position on a small team of data engineers with growth potential. The firm is looking for outstanding technical skills, strong attention to detail, and a desire to architect and build data platforms.

Location
Austin, TX / Yerevan, Armenia (in-office requirement)

Key Responsibilities

• Work directly with Portfolio Managers and Quantitative Developers: turn their requirements into datasets and pipelines, and be the person who knows every nuance of the data they trade on.

• Design and onboard new data sources into our warehouse; improve the robustness, speed, and scalability of our systems; manage data entitlements.

• Build automated systems for data cleansing, anomaly detection, monitoring, and alerting, bad data must never reach a strategy.

• Evaluate new tools and technologies for organizing, querying, and streaming large datasets, and when nothing on the market fits, build it. That's how the bitemporal store happened.

• Support the production data warehouse the firm depends on.

• Develop and maintain vendor relationships aligned with our business objectives.

What we're building right now

• A bitemporal data store, designed and written in-house from scratch. Every dataset answers both "what did we know then?" and "what do we know now?", which is what lets researchers trust a backtest.

• A Python 3.14 migration of a large, long-lived codebase.

• Adoption of the latest Apache Airflow: writing DAGs for the thousands of jobs moving off cron.

• Pipelines for market data and alternative data: everything from exchange feeds to weather.

• Real-time delivery: alternative data flows straight into strategies' live feeds. Pipelines you build sit in the trading path.

• CI/CD for all of it, in GitHub Actions.

Our Stack

Python and Java · Apache Airflow · Slurm · NATS · PostgreSQL · MongoDB · S3 · NFS · GitHub Actions

Basic Requirements

• Proficiency in Python and Unix/Linux for data manipulation, scripting, and automation.

• Strong SQL, including query optimization and performance tuning, and familiarity with NoSQL.

• A solid grasp of data modeling: normalization and denormalization, and the judgment to know when each applies.

Nice to have

• Financial industry experience or internships.

• Java (part of our platform is written in it).

• Experience with on-premises data infrastructure.

• Familiarity with a cloud platform (AWS or GCP).

• Apache Airflow or similar workflow orchestration tools.

Benefits

• Health, visual and dental insurance

• Flexible sick time policy