Ines Montani

Taking back control of your AI development

Sat, 06 Jun 2026 00:00:00 -0400

Short presentation from the Feminist AI unconference at PyData London, covering the current focus of my work.

Efficient data development: Optimizing annotation workflows

Tue, 24 Feb 2026 00:00:00 -0500

Tips and advice for how to build efficient human-in-the-loop data development workflows, break down business problems into actionable annotation steps and make the most of automation and model assistance. All examples are inspired by real use cases. Blog post: https://explosion.ai/blog/optimizing-annotation-workflows

PDFs bezwingen: Dokumentenanalyse über reinen Text hinaus

Wed, 10 Dec 2025 00:00:00 -0500

Talk for Bitkom e.V. AK Legal Tech (Berlin, 10.12.2025) NLP und Data Science könnten so einfach sein, wenn all unsere Daten als sauberer, reiner Text vorlägen. Doch in der Praxis sind sie meist versteckt in PDFs, Word-Dokumenten, Scans und anderen Formaten, deren Verarbeitung sich als Albtraum erwiesen hat. In diesem Talk präsentiere ich einen neuen, modularen Ansatz für die Entwicklung von robusten Systemen für Dokumentenanalyse mit Hilfe von modernsten Modellen und dem großartigen Python-Ökosystem. Ich zeige, wie wir von PDFs zu strukturierten Daten gelangen und sogar vollständig benutzerdefinierte Informationsextraktions-Pipelines für spezifische Anwendungsfälle aus der Praxis erstellen können. Für die praktischen Beispiele verwende ich spaCy, sowie die Bibliothek und Layoutanalysemodelle von Docling. Ich thematisiere außerdem optische Zeichenerkennung (OCR) für bildbasierten Text, die Verwendung von bewährten NLP-Techniken, und Strategien zur Erstellung von Trainings- und Evaluationsdaten für Informationsextraktionsaufgaben wie Textklassifizierung und Entitätserkennung anhand von PDFs und anderen Dokumenten. Blogpost: https://explosion.ai/blog/pdfs-nlp-structured-data

HU Berlin: Industrial-Strength Natural Language Processing with spaCy and Prodigy

Tue, 02 Dec 2025 00:00:00 -0500

Guest lecture for "Deep Learning and Natural Language Processing" at Humboldt University Berlin.

Applied NLP in the Age of Generative AI: Future-Proof Strategies for Banking and Finance

Thu, 05 Jun 2025 00:00:00 -0400

Large Language Models (LLMs) and in-context learning have introduced a new paradigm for developing natural language understanding systems: prompts are all you need! Prototyping has never been easier, but not all prototypes give a smooth path to production. Many new ideas that are emerging also challenge existing workflows in industry that require modularity, transparency and data privacy. In this talk, I'll share the most important lessons we've learned from solving real-world information extraction problems in industry, and show you a new approach and mindset for building modular and future-proof NLP pipelines in-house.

KI zwischen Freiheit und Kontrolle: The AI Revolution Will Not Be Monopolized

Tue, 11 Mar 2025 00:00:00 -0400

Generative KI und Large Language Models (LLMs) stellen Unternehmen, die auf Transparenz, Modularität und Privatsphäre setzen, vor neue Herausforderungen. Auf der einen Seite wartet enormes Potenzial, auf der anderen muss man das erst einmal finden, irgendwo zwischen spannenden neuen Tools, FOMO und Tech-Bros, die immer neue Revolutionen anpreisen. Wie müssen wir uns den Einsatz der neuen Technologien in der Praxis tatsächlich vorstellen? Und steuern wir dabei auf eine Black-Box-Ära zu, mit immer größeren Modellen hinter undurchsichtigen APIs, kontrolliert von Big-Tech-Monopolen? Talk for data:unplugged

Künstliche Intelligenz: Technologie der Zukunft – und warum Open Source die Karten neu mischt

Thu, 06 Mar 2025 00:00:00 -0500

Heise KI-Woche 2025: Wie sich KI in Zukunft in Europa entwickeln könnte - und warum Open Source die Karten neu mischt: Deep Seek hat vor einigen Wochen mit aufsehenerregenden Benchmarks den KI-Markt durcheinandergewirbelt. Ist das ein Vorbildmodell für Europa und wo entwickelt sich KI, vor allem Open Source, möglicherweise hin?

What the history of the web can teach us about the future of AI

Sat, 25 Jan 2025 00:00:00 -0500

Blog post: https://explosion.ai/blog/history-web-future-ai Video: https://www.youtube.com/live/kpocg6b89Fs?si=pmNN1kX5GJCe1vke&t=3840 Recent advancements in Generative AI are exciting, and will surely have a significant, yet uncertain impact on the future. Are we still going to need developers going forward, or will they be replaced by AI? Is Big Tech monopolizing the technology? And will we become entirely dependent on API providers, sacrificing the spirit of open-source software and data privacy? I believe there is a lot we can learn from another groundbreaking technology: the web. In this talk, I'll show you what the history of the web can teach us about the future of artificial intelligence, and what this means for developers, models, open source and regulation.

Reality is not an End-to-End Prediction Problem: Applied NLP in the Age of Generative AI

Thu, 17 Oct 2024 00:00:00 -0400

Video: https://www.youtube.com/watch?v=K_Y9wvGjNKw Large Language Models (LLMs) and in-context learning have introduced a new paradigm for developing natural language understanding systems: prompts are all you need! Prototyping has never been easier, but not all prototypes give a smooth path to production. In this talk, I'll share the most important lessons we've learned from solving real-world information extraction problems in industry, and show you a new approach and mindset for designing robust and modular NLP pipelines in the age of Generative AI. Breaking down larger business problems into actionable machine learning tasks is one of the central challenges of applied natural language processing. I will walk you through example applications and practical solutions, and show you how to use LLMs to their fullest potential, how and where to integrate your custom business logic and how to maximize efficiency, transparency and data privacy.

Applied NLP with LLMs: Beyond Black-Box Monoliths

Wed, 09 Oct 2024 00:00:00 -0400

Large Language Models (LLMs) have enormous potential, but also challenge existing workflows in industry that require modularity, transparency and data privacy. In this talk, I'll show some practical solutions for using the latest state-of-the-art models in real-world applications and distilling their knowledge into smaller and faster components that you can run and maintain in-house. _________________________________________________ ▪️ Case Study #1: https://speakerdeck.com/inesmontani/workshop-half-hour-of-labeling-power-can-we-beat-gpt ▪️ Case Study #2: https://explosion.ai/blog/sp-global-commodities ▪️ Case Study #3: https://explosion.ai/blog/gitlab-support-insights

Lightning Talk: Beautiful Slides for Beginners

Thu, 29 Aug 2024 00:00:00 -0400

People often ask me how I make my slides, so here I’m sharing some of my not-so-secret secrets and beginner-friendly steps for how you can up your slides game for the upcoming conference season. 1. Design tips for beginners: https://ines.io/blog/beginners-guide-beautiful-slides-talks/ 2. Finding your aesthetic: https://ines.io/blog/beautiful-slides-talks-part-2-aesthetics/ 3. Presenting technical content https://ines.io/blog/beautiful-slides-talks-part-3-technical-content/

10 Years of Open Source: Navigating the Next AI Revolution

Wed, 28 Aug 2024 00:00:00 -0400

A lot has been happening in the field of AI and Natural Language Processing: there's endless excitement about new technologies, sobering post-hype hangovers and also uncertainty about where the field is heading next. In this talk, I'll share the most important lessons we've learned in 10 years of working on open-source software, our core philosophies that helped us adapt to an ever-changing AI landscape and why open source and interoperability still wins over black-box, proprietary APIs.

Towards Structured Data: LLMs from Prototype to Production

Wed, 12 Jun 2024 00:00:00 -0400

Large Language Models (LLMs) have enormous potential, but also challenge existing workflows in industry that require modularity, transparency, data privacy and structured data. In this talk, I'll present pragmatic and practical approaches for how to use LLMs beyond just chat bots, how to ship more successful NLP projects from prototype to production and how to use the latest state-of-the-art models in real-world applications and distilling their knowledge into smaller and faster components that can be run and maintained in-house.

spaCy meets LLMs: Using Generative AI for Structured Data

Tue, 11 Jun 2024 00:00:00 -0400

Large Language Models (LLMs) have enormous potential, but also challenge existing workflows in industry that require modularity, transparency and structured data. In this talk, I'll present pragmatic and practical approaches for how to use the latest generative models beyond just chat bots. I'll dive deeper into spaCy's LLM integration, which lets you plug in open-source and proprietary models and provides a robust framework for extracting structured information from text, distilling large models into smaller task-specific components, and closing the gap between prototype and production.

The AI Revolution Will Not Be Monopolized: Behind the scenes

Sun, 21 Apr 2024 00:00:00 -0400

A more in-depth look at the concepts and ideas behind my talk "The AI Revolution Will Not Be Monopolized: How open-source beats economies of scale, even for LLMs", including academic literature, related experiments and preliminary results for distilled task-specific models.

The AI Revolution Will Not Be Monopolized: How open-source beats economies of scale, even for LLMs (QCon London)

Mon, 08 Apr 2024 00:00:00 -0400

With the latest advancements in Natural Language Processing and Large Language Models (LLMs), and big companies like OpenAI dominating the space, many people wonder: Are we heading further into a black box era with larger and larger models, obscured behind APIs controlled by big tech monopolies? I don’t think so, and in this talk, I’ll show you why. I’ll dive deeper into the open-source model ecosystem, some common misconceptions about use cases for LLMs in industry, practical real-world examples and how basic principles of software development such as modularity, testability and flexibility still apply. LLMs are a great new tool in our toolkits, but the end goal remains to create a system that does what you want it to do. Explicit is still better than implicit, and composable building blocks still beat huge black boxes. As ideas develop, we’re seeing more and more ways to use compute efficiently, producing AI systems that are cheaper to run and easier to control. In this talk, I'll share some practical approaches that you can apply today. If you’re trying to build a system that does a particular thing, you don’t need to transform your request into arbitrary language and call into the largest model that understands arbitrary language the best. The people developing those models are telling that story, but the rest of us aren’t obliged to believe them.

Workshop: Half hour of labeling power: Can we beat GPT?

Wed, 01 Nov 2023 00:00:00 -0400

Video: https://www.youtube.com/watch?v=Ta45SfbZNcM Large Language Models (LLMs) offer a lot of value for modern NLP and can typically achieve surprisingly good accuracy on predictive NLP tasks with a reasonably structured prompt and pretty much no labelled examples. But can we do even better than that? It’s much more effective to use LLMs to create classifiers, instead of using them as classifiers. By using LLMs to assist with annotation, we can quickly create labelled data and systems that are much faster and much more accurate than using LLM prompts alone. In this workshop, we'll show you how to use LLMs at development time to create high-quality datasets and train specific, smaller, private and more accurate fine-tuned models for your business problems.

Large Language Models: From Prototype to Production (EuroPython keynote)

Wed, 19 Jul 2023 00:00:00 -0400

Large Language Models (LLMs) have shown some impressive capabilities and their impact is the topic of the moment. What will the future look like? Are we going to only talk to bots? Will prompting replace programming? Or are we just hyping up unreliable parrots and burning money? In this talk, I'll present visions for NLP in the age of LLMs and a pragmatic, practical approach for how to use Large Language Models to ship more successful NLP projects from prototype to production today. Video: https://www.youtube.com/watch?v=ZjjgMiCU8s4 Twitter: https://twitter.com/_inesmontani/status/1681700743693172738 LinkedIn: https://www.linkedin.com/posts/inesmontani_nlp-llm-llms-activity-7087478372418625536-3VDo