Driving Industrial Innovation with AI at Scale
summary
ABB and Adastra build a pipeline on AWS that pulls a BOM from the e‑commerce platform, parses 1,000‑page technical publications with Textract, labels tables/diagrams, vectorizes content into metadata, then uses Bedrock LLMs to retrieve and summarize only project‑relevant sections into an ad hoc PDF. The result shrinks manuals to 50–100 pages, runs automatically per order, starts with one product line and is designed to scale across 650,000 products and $10B+ in annual e‑commerce revenue.
AI-written summary · about
chapters (7)
- 0:00Introduction and welcome
- 2:28ABB Electrification overview
- 4:35Partnership and ecosystem
- 6:03Problem with project documentation
- 7:27Solution architecture explained
- 11:41Results and benefits
- 13:06Scalability and future plans
related talks
32mUsing AI to power a circular economy for aerospace
43mImplementing AI Document Processing at Scale: Syngenta
01h 30mDriving growth with innovative digital products
20mExplore what’s possible with new AWS generative AI services
30mIntelligent document processing: Extract data & insights from documents
35mContinental Resources, PwC leverage generative AI to optimize oil & gas ops
31mThe Fast and the Curious: Harnessing AI in Financial Data Extraction at Scale
32mBeyond Bolt-On AI: A New Way to Build Software