Senior Crypto Data Engineer (Global-Remote-Non-US)

�

� Description

• • Architect and own the end-to-end big-data ingestion pipeline that feeds Token Metrics’ AI crypto-indices, price-prediction models and on-chain analytics consumed by 50 k+ investors worldwide. You will source, clean and structure billions of rows of market, social, on-chain and alternative data arriving via REST & WebSocket APIs, CSV dumps, JSON blobs, Parquet files and raw blockchain nodes.
• • Build a petabyte-scale Data Lake on AWS S3 and a query-optimized Data Warehouse in Snowflake, designing star and snowflake schemas that let quants run sub-second aggregations across 3 000+ tokens and 200+ exchanges without blowing the budget.
• • Implement near-real-time streaming jobs in Python (PySpark / Kafka / Kinesis) that normalise tick-level trades, order-book snapshots and gas fees, achieving ≤ 2 min latency from exchange to model while guaranteeing exactly-once semantics.
• • Create reusable PySpark & DBT frameworks for feature engineering (volatility regimes, whale-flow metrics, social-sentiment scores) so Data Scientists can iterate in hours instead of weeks; expose the features through low-latency REST and GraphQL endpoints.
• • Automate data-validation & reconciliation pipelines that compare exchange candles, detect gaps, quantify drift and self-heal by back-filling from redundant sources; maintain 99.9 % data-quality SLA and publish daily DQ dashboards for stakeholders.
• • Optimise storage and compute costs through intelligent partitioning, Z-order clustering, auto-scaling Snowflake warehouses and spot-instance strategies, cutting last year’s AWS bill by 30 % while doubling data volume.
• • Champion software-engineering best practices: GitFlow, unit/integration tests > 90 % coverage, CI/CD via GitHub Actions, containerised micro-services (Docker + EKS), infrastructure-as-code (Terraform) and peer-reviewed pull requests.
• • Collaborate cross-functionally with quant researchers, product managers and frontend engineers to translate vague business questions (“Which NFT wash-trading signals predict a 20 % price drop?”) into concrete data products delivered in agile two-week sprints.
• • Mentor junior engineers, conduct brown-bag sessions on modern data-stack tools and contribute to the company-wide data-governance council defining security, privacy and token-classification standards.
• • Stay ahead of the curve: evaluate new crypto data providers, experiment with Rust-based indexing, Delta Lake, dbt Mesh or DuckDB for ad-hoc analytics and present findings that shape next-quarter roadmap.

� Requirements

• • 3+ years production-grade Python, Java or Scala development and 3+ years writing complex SQL on TB-scale datasets; expert in both OLTP and OLAP query tuning.
• • Proven experience designing and operating a Data Lake (S3) and Data Warehouse (Snowflake, Redshift or BigQuery) including schema design, partitioning, clustering and cost governance.
• • Hands-on with distributed data processing (Spark, PySpark, Hive or Flink) and streaming frameworks (Kafka, Kinesis, Pulsar); comfortable deploying on AWS (EMR, Glue, Lambda, EKS).
• • Solid grasp of data-modelling techniques (Kimball star schema, data vault, slowly changing dimensions) and data-quality frameworks (Great Expectations, Deequ, Monte Carlo).
• • Nice-to-have: exposure to blockchain data (BigQuery crypto-public-data, Ethereum ETL, Dune, Flipside), NoSQL stores (MongoDB, Cassandra), Airflow or Dagster orchestration and a Master’s in CS, Data Engineering or quantitative field.

️ Benefits

• • Work 100 % remotely from any non-US location with flexible hours and asynchronous culture; we care about outcomes, not seat-time.
• • Competitive USD-based salary paid monthly in fiat or crypto (USDC/BTC) plus performance-based token allocation that vests over two years.
• • Annual US$ 2 000 learning stipend for courses, conferences or certifications (AWS, Spark, dbt, blockchain analytics) and free Token Metrics premium subscription.
• • 30 days paid time off, local public holidays and a quarterly “ recharge week ” where the whole company powers down.