Skip to content
Enterprise Data Connectors

Your warehouse, pre-enriched

Pull compound libraries from Snowflake, Databricks, BigQuery, or Supabase. Run ADMET predictions and compliance screening across every row. Push enriched results back. Schema discovery and field mapping included - without ETL engineering or new infrastructure.

4
Warehouse connectors
69
Enrichment tools
Bi-directional
Pull → Enrich → Push
Zero ETL
No pipeline code

How it works

Connect, enrich, push back

Snowflake, Databricks, BigQuery, Supabase - connect with credentials or OAuth. Schema discovery introspects your tables and auto-detects the SMILES column. Choose processing tools, preview the cost, and run. Results land back in your warehouse - without writing a single line of pipeline code.

Terminal - pull_from_source

Every row, fully characterized

ADMET predictions, regulatory compliance, drug-likeness, and structural alerts computed for every compound in your library. 31 ML models and 8 regulatory jurisdictions - applied per-row, not per-batch. Each row gets its own complete molecular profile.

ADMET Radar - Aspirin

CC(=O)Oc1ccccc1C(=O)O

AbsorptionDistributionMetabolismExcretionToxicity

85%

Abs

72%

Dis

91%

Met

78%

Exc

64%

Tox

Large libraries, tracked

Enrichment jobs run async for large libraries - up to 10,000 rows per pipeline. Track progress from the jobs dashboard or ask your AI. Credit cost previewed before execution. Full per-molecule audit trail for every pipeline run.

NovoMCP

Compute Jobs

Auto-refresh

Active

2

Completed today

18

Avg runtime

7m

run_molecular_dynamicsRunning
md_7f3a8c
Started 8m ago67% · ~4m
dock_moleculesCompleted
dock_2e91b4
Started 12m ago100% · done
predict_structureRunning
struct_a4f2d1
Started 3m ago34% · ~6m
screen_libraryCompleted
screen_c8e5f9
Started 1h ago100% · done
The funnel

The enrichment workflow

01

Connect

Add your Snowflake, Databricks, BigQuery, or Supabase credentials. OAuth for BigQuery - no service accounts needed.

02

Discover

Schema introspection finds your tables, columns, and types. The SMILES column is auto-detected. Preview row count and sample data.

03

Preview cost

Choose processing tools - ADMET, compliance, properties, optimization. See the exact credit cost before committing. The estimate doubles as a 21 CFR Part 11 audit artifact.

04

Enrich

Run the pipeline. Every row processed through your selected tools. Results written row-by-row with per-molecule audit logging.

05

Push back

Enriched results land in your destination table - the same warehouse, a different schema, or a new connector entirely.

Capabilities

Connector capabilities

Available on Scale and Enterprise tiers.

Snowflake

Account identifier, username, password. Specify warehouse, database, and schema. Full read/write access.

Databricks

Personal access token with workspace URL. Target catalog and schema. Unity Catalog compatible.

BigQuery

OAuth - click "Connect with Google" and authorize. No API keys or service accounts needed. Tokens refresh automatically.

Supabase

Connection string from your project dashboard. Service role key for full access. PostgreSQL-compatible.

Schema discovery

Introspect tables, columns, and types. Auto-detect SMILES columns. Normalize to 5 standard types. Filter by name pattern.

Audit trail

Per-molecule processing log for every pipeline run. SMILES validation, tool results, dispositions, exclusion reasons. 21 CFR Part 11 compliant.

Research preview

Connect your warehouse

Pull compounds, enrich with pre-computed ADMET and FAVES compliance, push results back. Set up in under five minutes.