⚡ Lightning AI Meetup: Cracking OCR
Join us in London on 3 September for an OCR-themed meetup hosted by Lightning AI, the team behind PyTorch Lightning.
Expect sharp talks and real conversations with developers and researchers turning unstructured images and scanned docs into usable data. If you touch invoices, PDFs, historical archives, receipts, forms, or any “this should be text but isn’t,” this night is for you.
🎙️ Speakers
Jonathan Bourne — Research Fellow, Centre for Advanced Spatial Analysis, UCL
Reading the Unreadable — OCR in the wild, from noisy scans to structured text
Marco Humbel — Honorary Research Fellow, Department of Information Studies, UCL
Transforming Collections into Data — Handwritten Text Recognition for Historic Documents
Sreekiran A R — Consultant, Data Science at LatentView Analytics
Building Trustworthy AI Solutions & Storytelling — From Data Wrangling to Decision Making
Exploring how structured architectures, clarity, and storytelling make AI solutions reliable, trustworthy, and impactful in real-world contexts.
🚀 Why Attend?
See practical pipelines for OCR on messy, low-quality inputs
Learn how to handle layout, handwriting, stamps, and tables
Get tips for evaluation, post-processing, and human-in-the-loop review
Connect with the London AI/ML community and fellow data wranglers
Enjoy free food, drinks, and good conversation 🍕🧃
📅 Schedule
6:00–6:30 PM: Registration & Networking
6:30–7:30 PM: Opening Remarks & Talks
7:30–8:30 PM: Food & Open Networking
👤 Who Should Attend?
ML engineers, data scientists, researchers, and builders working with OCR, document AI, RAG on PDFs, and information extraction. Also welcome: anyone dealing with scans, forms, and unstructured documents who wants better tooling.
📩 Questions?
Contact Kaya Woods, Developer & Community Program Manager, at [email protected].
See you there! ⚡️