Data Scientist, Blockchain Intelligence (Production ML)

Merkle Science; New York (NY); On-site; USD 140,000 - 180,000; Full time

Benefits offered by this job

Flexible time off

Learning & development

Equity opportunity

Job summary

Merkle Science is seeking a data scientist who can ship production-grade data pipelines for on-chain analytics. You will turn messy hypotheses into working, auditable pipelines that scale across billions of transactions and multiple chains.

Collaborating with attribution and clustering leads, you will design heuristics, own end-to-end data processing from ingest to validation, and move quickly from prototype to production-ready solutions.

Qualifications

• 4+ years shipping data science or data engineering systems that actually shipped.
• Strong Python and SQL; comfortable with large datasets and joins, dedup, and skew at scale.
• Solid grasp of clustering, graph analysis, or entity resolution with validation.
• Ability to reason about precision vs. coverage and defend metrics.
• Self-directed: scope ambiguous problems, fetch data, and drive results.

Responsibilities

• Design, test, and ship clustering and attribution heuristics with measurable precision.
• Own end-to-end data: pull, clean, join, and model large on-chain datasets.
• Build and maintain pipelines from notebook to production with backfills and validation.
• Investigate edge cases ( mixers, bridges, exchange wallets ) and translate findings into logic.
• Partner with investigations and product to define correctness and benchmark against ground truth.
• Prototype quickly, then harden what works.

Skills

Python

SQL

Databricks

Graph analysis

Entity resolution

Data pipelines

Big data

Tools

Databricks

Kafka

TigerGraph

ClickHouse

Git / GitHub

GCP

Job description

Merkle Science is seeking a data scientist who can ship production-grade data pipelines for on-chain analytics. You will turn messy hypotheses into working, auditable pipelines that scale across billions of transactions and multiple chains.

Collaborating with attribution and clustering leads, you will design heuristics, own end-to-end data processing from ingest to validation, and move quickly from prototype to production-ready solutions.