Sharing this data for research and AI development requires rigorous de-identification to ensure patient privacy — without compromising the ability to extract insights across time and modalities.

In this webinar, experts from John Snow Labs and Databricks will demonstrate an end-to-end solution for automating the de-identification and tokenization of medical data with regulatory-grade accuracy. You’ll learn how to:

• Automatically de-identify structured data, unstructured text, DICOM & JPEG images, whole-slide pathology images (SVS), and PDFs using John Snow Labs’ industry-leading software and AI models
• Apply patient tokenization to enable linking of de-identified data across modalities and time points
• Use Databricks to process and scale these capabilities across large, real-world datasets
• Support HIPAA, GDPR, and other regulatory requirements for privacy-preserving research

This session is ideal for data scientists, clinical researchers, compliance teams, and healthcare IT leaders working with multimodal patient data who want to enable longitudinal, privacy-compliant research at scale.

This webinar introduces the recently released Spark NLP 6.0, an Apache 2.0 licensed open-source Python library which enables you to analyze large amounts of multi-modal data for batch LLM inference or to prepare data for RAG & LLM solutions – privately, efficiently, and at no cost. The library can operate on a single machine or container, or scale natively on any Spark hardware without code changes. Spark NLP recently crossed 150M downloads and this new release adds supports 3 major new use cases:

• Support for ingesting and pre-processing PDF, Excel, PowerPoint, text and image files. Prepare, analyze, and ingest all files formats into a LLM / RAG solution using one unified pipeline.
• Visual language models! Multiple VLMs of different sizes & features are natively available as steps in processing pipelines, enabling you to extract facts and answers from images and visual PDF files.
• Extract structure, semantics, and metadata from unstructured and visual data in all file formats – using batch inference at scale.

Join to learn how to apply these new capabilities by walking through Python notebooks showcasing end-to-end scenarios.

Join us to explore the dual themes of agent evaluations and EvalOps in this comprehensive technical session. This webinar will focus on how to effectively integrate cost-effective LLM judges into AI systems, enhancing the precision and efficiency of agent decisions and overall LLM evaluations. You’ll learn about the strategic application of EvalOps and agent-specific evaluations, balancing accuracy, speed, and cost, backed by practical case studies.

Key Takeaways:
– Gain insights into both specific agent evaluations and general LLM evaluations through EvalOps.
– Discover strategies to optimize both agent performance and overall system evaluations.
– Examine case studies demonstrating successful implementations of agent evaluations and EvalOps.
– Learn how to maintain budget control while maximizing performance and accuracy.

This webinar is designed for ML engineers and data scientists looking to deepen their expertise in sophisticated AI model management.

Healthcare systems, payers, and medical societies invest massive effort to maintain evidence-based clinical guidelines for a variety of conditions. However, when patients are in the hospital, often clinicians just don’t have the time to research or read these guidelines, leading to major gaps in how consistently they are applied. Recent advances in Medical AI can shortcut this problem by automatically reading the full history of a given patient, finding the most recent and relevant guideline for their clinical history, and presenting it in context. This session will walk through a solution architecture for an end-to-end solution that does this, using a state-of-the-art healthcare-specific LLM, that can be deployed locally within an organization’s security perimeter to ensure privacy, compliance, and the ability to read organization-specific guideline documents. We’ll also show how to handle formatting of clinical guideline documents that are challenging to general-purpose LLMs like flowcharts, decision trees, and visual decision tables.

Implementing LLM “judges” in production settings can be a game-changer for evaluating AI behaviors, but it’s often more challenging than it appears. Many teams struggle with common pitfalls such as; high error rates and cost unpredictability to issues with latency and long-term maintenance. This webinar will break down the top 10 misconceptions around LLM judges, equipping you with the insights to avoid these challenges and build more reliable, production-ready evaluation systems.

Join us to learn:
• Key Misconceptions: Understand where teams often go wrong in deploying LLM judges.
• EvalOps Principles: Discover best practices and tools to operationalize evaluations effectively.
• Reliability in Production: Learn how to make evaluation outcomes systematic, scalable, and dependable.
• Use Cases & Value: Explore real-world applications where LLM judges add significant value and the tangible results they deliver.

Who Should Attend:
This webinar is ideal for machine learning engineers, data scientists, AI practitioners, and technical leaders seeking to enhance their approach to model evaluation in production environments.

Data integration has been an enormous challenge in healthcare for decades. Issues of data standardization, data quality, legacy formats, unstructured data, and semantic inconsistencies have made it hard to answer basic questions about how a hospital operates or what should be done next for a patient. Recent advances in Healthcare AI combine to transform this age-old problem – enabling you to automatically ingest large volumes of raw, multi-format, multi-modal, untrusted medical data into coherent longitudinal patient stories in an industry-standard format.

This webinar presents an integrated solution in action that uses John Snow Labs’ state-of-the-art Medical Language Models, healthcare-specific data preparation pipelines, and Text-to-OMOP question answering models running on Databricks’ secure, scalable, and compute-optimized AI platform. The solution takes in multi-modal data – structured (tabular), semi-structured (FHIR resources), and unstructured (free-text) – and generates an OMOP/OHDSI standard data model that:

– Builds a unified view of each patient over time.
– Builds this unified patient view from multi-modal source data.
– Reasons at the patient level.

We’ll then show how the resulting patient data model can then be used for either “AI” (building patient cohorts with natural language queries) or for “BI” (dashboards for patient risk scoring and quality measures), all from the same source of truth, with full explainability and traceability.

Synthetic data has emerged as a powerful tool for organizations looking to enhance their operational efficiency and reduce costs. By generating artificial data that mimics real-world data, companies can overcome limitations related to data scarcity, privacy concerns, and high costs associated with data collection and storage. This conversation will delve into the practical benefits and real-world applications of synthetic data, providing insights on how it can drive significant efficiency gains and cost savings.

Time series analysis is ubiquitous in applied data science because of the value it delivers. In order to do effective time series analysis, you need to know your tools well. Polars has excellent built-in time series support, and it’s also possible to extend it where necessary.

In this talk, we will cover:
– Basic built-in time series operations with Polars (e.g. “what’s the average number of sales per month?”).
– Advanced, custom time series operations, and how you can implement them as Polars plugins (e.g. business day arithmetic).
– What Polars can’t (yet!) do, and what we can do about it

Audience level:
Basic interest and knowledge of Python and data will be assumed, but no prior Polars experience is required.

—

This event is hosted by OpenTeams (https://www.openteams.com), the sponsor of REPOS, a platform that promotes sustainable funding for open source projects. To learn more, visit https://repos.openteams.com.

Have you ever considered pursuing a career in training? For our guest speaker, Reuven Lerner, this path was not obvious at first. With a degree in Computer Science, the expectation was a career focused on software engineering and architecture. However, after many years of teaching Python and pandas full-time, he has found both greater financial rewards and a deeper sense of fulfillment and excitement in his work. In this talk, he will share valuable insights into the world of corporate training, his journey into this field, and whether it might be a suitable path for you. The session will conclude with an open Q&A, providing an opportunity for attendees to ask questions and learn more.

The future of generative-AI is multimodal and yet, projects involving LLM and Large Vision Models in production are already struggling as teams try to scale up (and out) awe-inspiring Twitter demos. We’ll discuss what teams need to be doing to support the development and integration of multimodal models into product features.

Learn how the open-source Spark NLP library provides optimized and scalable LLM inference for high-volume text and image processing pipelines. This session dives into optimized LLM inference without the overhead of commercial APIs or extensive hardware setups. We will show live code examples and benchmarks comparing Spark NLP’s performance and cost-effectiveness against both commercial APIs and other open-source solutions.

Key Takeaways:

• Learn how to efficiently process millions of LLM interactions daily, circumventing the costs associated with traditional LLM deployments.
• Discover advanced methods for embedding LLM inference within existing data processing pipelines, enhancing throughput and reducing latency.
• Review benchmarks that compare Spark NLP’s speed and cost metrics relative to commercial and open-source alternatives.

Turning open-source projects into sustainable businesses is a crucial challenge for those in the tech industry. This AMA session explores practical strategies for building profitable business models within the open-source landscape. Travis and Peter will talk about various revenue streams such as support services, premium features, dual licensing, and partnerships. Participants will learn how to effectively communicate the value of open-source solutions, engage with communities for support and feedback, and balance open-source ethos with commercial objectives.

Join us and bring your questions for our speakers, Peter Wang from Anaconda and Travis Oliphant from OpenTeams, as they provide insights on transforming open-source initiatives into lasting ventures.

This talk presents new levels of accuracy that have very recently been achieved, on public and independently reproducible benchmarks, on the three most common use cases for language models in healthcare:

• Understanding clinical documents: Such as information extraction from clinical notes and reports; detecting entities, relationships, and medical codes; de-identification; and summarization.
• Reasoning about patients: Fusing information across multiple modalities (tabular data, free text, imaging, omics) to create a longitudinal view of each patient, including making reasonable inferences and explaining them.
• Answering medical questions: Answering medical licensing exam questions, biomedical research questions, and similar medical knowledge questions – accurately, without hallucinations, and while citing relevant sources.

On April 5th, 2024, the open-source community faced a significant challenge when a malicious backdoor was discovered in xz Utils, a popular data compression utility. This incident, triggered by the original maintainer’s burnout and subsequent manipulation by an attacker, underscores the vulnerabilities that arise from insufficient community support and the pressures faced by open-source maintainers.

This webinar presents key findings from the 2024 Generative AI in Healthcare Survey, conducted in February & March of 2024 by Gradient Flow to assess the key use cases, priorities, and concerns of professionals and technology leaders in this space. Topics covered:

• Current levels of adoption and budget allocation
• Types of language models being used
• Use cases for LLMs
• Priorities for evaluating LLMs and roadblocks
• LLM model enhancement strategies
• LLM testing for Responsible AI requirements

This webinar dives into the essence of effective leadership and the challenges inherent in translating leadership theory into practice. Acknowledging that most leadership behaviors are deeply ingrained and habitual, the session will explore the complexities of habit formation in a professional context. Drawing on new insights from cognitive sciences, the webinar will dissect why forming new leadership habits is a challenging endeavor and will outline strategies for successful habit creation. Tobi will illustrate this journey through three foundational leadership habits: Being Present, Sharing Context, and Setting Boundaries. Each habit will be discussed in detail, providing you with a roadmap for connecting with colleagues, sharing knowledge and perspectives, and establishing clear principles and boundaries. The session aimed to equip leaders with practical advice and actionable strategies to foster healthy leadership habits that can significantly impact their effectiveness and the well-being of their teams.

This webinar is designed to address one of the most pressing questions for creators and sales professionals in the open-source space: how to generate sustainable revenue from open-source products. It will explore various monetization strategies, discuss the challenges and opportunities inherent in open-source business models, and provide actionable insights for effectively selling and scaling open-source solutions.

Learn to enhance Retrieval Augmented Generation (RAG) pipelines in this webinar on John Snow Labs’ integrations with LangChain and HayStack. This session highlights the ability to retain your existing pipeline structure while upgrading its accuracy and scalability. Accuracy is improved thanks to customizable embedding collection and document splitting. Using Spark NLP’s optimized pipelines greatly improves scalability, runtime speed, and as a result cost.

Learn how these native integrations enable an easy transition to more effective methods, enhancing document ingestion from diverse sources without overhauling existing systems. Whether your goal is to enhance data privacy, optimize NLP & LLM accuracy, or scale your RAG applications to millions of documents, this webinar will equip you with the knowledge and tools to fully leverage John Snow Labs’ software to get it done. Join us to unlock the potential of your applications with the latest innovations in Generative AI, without departing from the familiar toolset of your current pipeline.

Explore the latest advancements in multimodal AI for extracting tabular data from visual documents. This session will delve into novel methods implemented in John Snow Labs’ Visual NLP library, which has significantly improved the accuracy of information extraction and question answering from tables in PDFs and image files.

The webinar will cover a range of practical applications, demonstrating how this technology is adept at handling complex documents such as financial disclosures, clinical trial results, insurance rates, lab scores, and academic research. The focus will be zero-shot models, where the AI model directly interprets and responds to queries from source images, eliminating the need for specialized training or tuning.

We’ll also cover Visual NLP capabilities that have been specifically designed to enhance table extraction quality, especially in challenging cases like multi-line cells or borderless tables. We’ll discuss the technical underpinnings of this feature, including the integration of computer vision and object character recognition for detecting tables and individual cells within them. We’ll touch upon how that extends to support for tables with custom borders, dark & noisy backgrounds, uncommon table layouts, multilingual text, and international number & currency formats.

In this webinar, we’ll be exploring the intricate process of effectively managing and scaling multiple LLM applications. Our focus will extend beyond the basic use of notebook-based methods, diving into more robust and advanced strategies. From the initial development stages right through to deployment, this webinar is tailored to provide a comprehensive understanding of the complexities and nuances involved in operationalizing LLMs at scale.

Key Topics We’ll Cover:

1. Advanced Management Techniques: Step beyond traditional approaches with insights into state-of-the-art strategies for managing and scaling LLM applications.
2. Operationalization Challenges: Uncover the critical aspects often missed in the journey from development to deployment, emphasizing the importance of ongoing maintenance and monitoring for sustained success.
3. Scalability in Practice: Equip yourself with practical solutions to effectively scale LLM applications, ensuring they meet the evolving demands and technological trends in today’s fast-paced environment.