Drug Discovery Intelligence Workflow (n8n): Claude, Pinecone, Airtable & Slack

Turn daily drug-discovery signals into structured intelligence—automatically

The Drug Discovery Intelligence Workflow (n8n): Claude, Pinecone, Airtable & Slack runs every day to pull new research and development updates from major sources, extracts structured R&D intelligence with Anthropic Claude, indexes results in Pinecone, saves records to Airtable, and notifies you in Slack when high-value discoveries appear.

What this workflow does

• Runs daily (6am) or manually and loads your configured search keywords, watch-target list, thresholds, and destination settings.
• Collects new items from PubMed, bioRxiv/medRxiv, a configured patents API, and ClinicalTrials.gov.
• Normalizes, deduplicates, and processes only new documents by comparing against workflow static data from previous runs.
• For each new document, sends the text to Anthropic Claude to extract structured fields such as targets, molecules, mechanism of action, development stage, competing organizations, and a summary.
• Scores strategic relevance using watch-target matching and first-in-class signals.
• Creates embeddings and indexes into Pinecone (via Voyage AI embedding) with metadata for retrieval and downstream use.
• Upserts intelligence records into Airtable and posts an immediate Slack alert for high-scoring discoveries.
• Posts a Slack run digest at completion, including a completion note and cumulative indexed-document count.

Use cases

• Biotech/SaaS teams tracking targets and first-in-class signals without manually scanning PubMed, preprints, patents, and trials.
• Automation engineers building a drug discovery intelligence pipeline that converts articles into searchable vector data.
• Operators who want daily Slack alerts and a structured Airtable database for triage.

Technical details

• Built in n8n using nodes including if, set, code, merge, Slack, and Airtable.
• Integrations: Anthropic Claude for extraction, Pinecone for vector indexing, and Slack for alerts and run digests.
• Uses workflow static data for deduplication across runs.

Result: a practical, production-style n8n automation workflow that continuously transforms drug discovery sources into actionable intelligence—ready for search, review, and decision-making.