Data Scientist
Delta System & Software, Inc.
•
Berkeley Heights, NJ
0
0
0
0
Mid-Senior level
Posted April 2, 2026
Job link
Thinking about this job
Responsibilities
Commitments
Responsibilities
- Design and evolve graph data models for financial events, entities, and relationships (accounts, payments, invoices, trades, counter parties, ownership, etc.).
- Translate business questions into graph queries and features (traversals, communities, centrality, paths, temporal patterns).
- Build data pipelines for ingestion, cleaning, labelling, and feature engineering, including entity resolution and relationship extraction where needed.
- Develop and validate statistical/ML models (risk scoring, anomaly detection, fraud patterns, forecasting, classification).
- Create event-driven analytics using strong time semantics (event ordering, windows, causality assumptions, life-cycle states).
- Partner with engineering to productive models: batch + near-real-time scoring, monitoring, drift checks, and reproducible experiments.
- Communicate findings clearly via notebooks, dashboards, and concise write-ups.
Commitments
Role: Data Scientist — Financial Events & Graph Analytics (Graph DB / REA a Plus) Location: Berkeley Heights, NJ (53 Days) and Princeton, NJ(2 Days) (based on client schedule Duration : Permanent Type : Full-time Role summary We’re hiring a Data Scientist to model and analyse financial events and entity relationships using graph data.
Not Met Priorities
What still needs stronger evidence
Requirements
- Familiarity with REA (Resources–Events–Agents) accounting/event modeling is a plus.
- Communicate findings clearly via notebooks, dashboards, and concise write-ups.
- Strong foundation in statistics + machine learning (evaluation, leakage prevention, bias checks, calibration, experimentation).
- Hands-on experience with Graph DBs and graph concepts:
- Schema/design: node/edge types, properties, constraints, indexing, cardinality, temporal modeling
- Querying: Cypher (Neo4j) and/or Gremlin/SPARQL
- Graph algorithms: Page Rank, betweenness, connected components, community detection, similarity
- Strong Python for DS (pandas, numpy, scikit-learn; comfort writing production-ready code).
- Solid data engineering basics: SQL, ETL, data quality checks, versioning, reproducibility.
- Ability to explain technical results to non-technical stakeholders.
- Financial data and event modeling: payments, reconciliation, ledgers, trades, positions, KYC/AML signals, counterparty networks.
- Understanding of financial events and workflows (authorization → capture → settlement, invoice → payment → reconciliation, trade lifecycle, etc.).
- REA (Resources–Events–Agents) modeling and/or accounting event-sourcing concepts is a strong plus.
- Entity resolution / record linkage; graph-based identity resolution.
- Experience with cloud data stacks (GCP/AWS), orchestration (Airflow/Prefect), and model serving.
- Knowledge of governance/security patterns for sensitive financial data
Preferred Skills
- Familiarity with REA (Resources–Events–Agents) accounting/event modeling is a plus.
- Domain experience (preferred)
- Financial data and event modeling: payments, reconciliation, ledgers, trades, positions, KYC/AML signals, counterparty networks.
- Understanding of financial events and workflows (authorization → capture → settlement, invoice → payment → reconciliation, trade lifecycle, etc.).
- REA (Resources–Events–Agents) modeling and/or accounting event-sourcing concepts is a strong plus.
- Entity resolution / record linkage; graph-based identity resolution.
- NLP for event extraction from unstructured text (contracts, filings, invoices).
- Experience with cloud data stacks (GCP/AWS), orchestration (Airflow/Prefect), and model serving.
- Knowledge of governance/security patterns for sensitive financial data
Role: Data Scientist — Financial Events & Graph Analytics (Graph DB / REA a Plus) Location: Berkeley Heights, NJ (53 Days) and Princeton, NJ(2 Days) (based on client schedule Duration : Permanent Type : Full-time Role summary We’re hiring a Data Scientist to model and analyse financial events and entity relationships using graph data. You’ll work with engineers and stakeholders to design graph schemas, build analytical pipelines, and deliver insights/products such as risk signals, anomaly detection, entity resolution, and event-driven intelligence. Familiarity with REA (Resources–Events–Agents) accounting/event modeling is a plus. What you’ll do
Design and evolve graph data models for financial events, entities, and relationships (accounts, payments, invoices, trades, counter parties, ownership, etc.).
Translate business questions into graph queries and features (traversals, communities, centrality, paths, temporal patterns).
Build data pipelines for ingestion, cleaning, labelling, and feature engineering, including entity resolution and relationship extraction where needed.
Develop and validate statistical/ML models (risk scoring, anomaly detection, fraud patterns, forecasting, classification).
Create event-driven analytics using strong time semantics (event ordering, windows, causality assumptions, life-cycle states).
Partner with engineering to productive models: batch + near-real-time scoring, monitoring, drift checks, and reproducible experiments.
Communicate findings clearly via notebooks, dashboards, and concise write-ups. Must-have skills
Strong foundation in statistics + machine learning (evaluation, leakage prevention, bias checks, calibration, experimentation).
Hands-on experience with Graph DBs and graph concepts:
Schema/design: node/edge types, properties, constraints, indexing, cardinality, temporal modeling
Querying: Cypher (Neo4j) and/or Gremlin/SPARQL
Graph algorithms: Page Rank, betweenness, connected components, community detection, similarity
Strong Python for DS (pandas, numpy, scikit-learn; comfort writing production-ready code).
Solid data engineering basics: SQL, ETL, data quality checks, versioning, reproducibility.
Ability to explain technical results to non-technical stakeholders. Domain experience (preferred)
Financial data and event modeling: payments, reconciliation, ledgers, trades, positions, KYC/AML signals, counterparty networks.
Understanding of financial events and workflows (authorization → capture → settlement, invoice → payment → reconciliation, trade lifecycle, etc.).
REA (Resources–Events–Agents) modeling and/or accounting event-sourcing concepts is a strong plus. Nice-to-have
Entity resolution / record linkage; graph-based identity resolution.
NLP for event extraction from unstructured text (contracts, filings, invoices).
Experience with cloud data stacks (GCP/AWS), orchestration (Airflow/Prefect), and model serving. Knowledge of governance/security patterns for sensitive financial data
Design and evolve graph data models for financial events, entities, and relationships (accounts, payments, invoices, trades, counter parties, ownership, etc.).
Translate business questions into graph queries and features (traversals, communities, centrality, paths, temporal patterns).
Build data pipelines for ingestion, cleaning, labelling, and feature engineering, including entity resolution and relationship extraction where needed.
Develop and validate statistical/ML models (risk scoring, anomaly detection, fraud patterns, forecasting, classification).
Create event-driven analytics using strong time semantics (event ordering, windows, causality assumptions, life-cycle states).
Partner with engineering to productive models: batch + near-real-time scoring, monitoring, drift checks, and reproducible experiments.
Communicate findings clearly via notebooks, dashboards, and concise write-ups. Must-have skills
Strong foundation in statistics + machine learning (evaluation, leakage prevention, bias checks, calibration, experimentation).
Hands-on experience with Graph DBs and graph concepts:
Schema/design: node/edge types, properties, constraints, indexing, cardinality, temporal modeling
Querying: Cypher (Neo4j) and/or Gremlin/SPARQL
Graph algorithms: Page Rank, betweenness, connected components, community detection, similarity
Strong Python for DS (pandas, numpy, scikit-learn; comfort writing production-ready code).
Solid data engineering basics: SQL, ETL, data quality checks, versioning, reproducibility.
Ability to explain technical results to non-technical stakeholders. Domain experience (preferred)
Financial data and event modeling: payments, reconciliation, ledgers, trades, positions, KYC/AML signals, counterparty networks.
Understanding of financial events and workflows (authorization → capture → settlement, invoice → payment → reconciliation, trade lifecycle, etc.).
REA (Resources–Events–Agents) modeling and/or accounting event-sourcing concepts is a strong plus. Nice-to-have
Entity resolution / record linkage; graph-based identity resolution.
NLP for event extraction from unstructured text (contracts, filings, invoices).
Experience with cloud data stacks (GCP/AWS), orchestration (Airflow/Prefect), and model serving. Knowledge of governance/security patterns for sensitive financial data