A Bit of AI Episode 7

2021-04-22 · Source: Explosion · Developer tools and consulting for AI, Machine Learning and NLP - Explosion.ai · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering, Data Science & Analytics · Depth: Novice, extended

Summary

A Bit of AI Episode 7 features Ines Montani, co-founder of Explosion, discussing her company's open-source NLP library, spaCy, and the annotation tool, Prodigy. She details her remote work routine, the development of spaCy 3.0 with new features for complex pipeline training and workflow management, and Prodigy Teams, a private cloud SaaS version. Montani explains that project inspiration comes from solving user problems and anticipating trends, highlighting how "raising a client round" funded Explosion. She also shares her non-traditional path into AI, combining linguistics and programming, and expresses frustration over misinformation and unethical PR surrounding AI's capabilities.

Key takeaway

For AI/ML engineers and data scientists building NLP solutions, Explosion's tools offer robust capabilities. You should explore spaCy 3.0 for enhanced workflow management and state-of-the-art models, and consider Prodigy Teams for privacy-preserving, collaborative data annotation. This approach streamlines development and leverages transfer learning effectively, reducing reliance on massive datasets.

Key insights

AI developer tools like spaCy and Prodigy are crucial for democratizing and streamlining NLP solution development.

Principles

• "Raising a client round" can bootstrap a company.
• Data labeling is an iterative development process.
• Transfer learning reduces data annotation needs.

Method

Explosion's project development involves solving user problems, anticipating trends like transfer learning, and using "raising a client round" for market analysis and funding.

In practice

• Explore spaCy for advanced NLP pipelines.
• Utilize Prodigy for efficient ML data annotation.
• Consider a "client round" for startup funding.

Topics

• Natural Language Processing
• spaCy
• Prodigy
• Machine Learning Tools
• Data Annotation
• Transfer Learning
• Ethical AI

Best for: NLP Engineer, AI Engineer, Machine Learning Engineer, AI Student

Related on AIssential

• PyCon Colombia Speaker Interview — Production-ready NLP tools and quality data annotation are crucial, with domain expertise being more vital than ML for its own sake.
• Explosion in 2022: Our Year in Review — Explosion significantly enhanced its NLP tools, spaCy and Prodigy, in 2022 with new components, speed, and language support.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Explosion · Developer tools and consulting for AI, Machine Learning and NLP - Explosion.ai.