Giskard

Featured in the latest Starlet issue by #StarHistory! 🌟 Star History is the #1 GitHub star history graph on the web, and tracks trending open-source projects and their growth over time. In their latest issue, you'll discover how our open-source library for testing LLMs helps identifying risks, such as hallucinations or prompt injections. It saves developers time by automating the evaluation process and money by avoiding costly AI incidents. A big thank you to Star History for featuring us! 🙌 📖 Read the article: https://gisk.ar/4eIbT8t #LLMeval #opensource #GitHub #AITesting

🚨 Have you taken our free course on #RedTeaming LLM applications yet? 🏴☠️🚩 Developed in collaboration with DeepLearning.AI, is designed to help you ensure the safety and robustness of your LLM apps. You'll learn how to: ⬩ Identify vulnerabilities in your LLM apps ⬩ Conduct red teaming exercises to test your systems ⬩ Explore real-world prompt injection attacks on chatbots ⬩ Understand and address LLM failure modes ⬩ Gain hands-on experience with both manual and automated red teaming methods 👉 Enroll for free: https://gisk.ar/4bmAaOg #LLMs #RAG #AISafety #AIRedTeaming

Recognized as a top AI innovator in #MAD50! 🤯 Our CEO, Alex Combessie 🐢 has been recognized in MAD50 ranking! MAD50, organized by Maddyness and partnered with Banque Richelieu, spotlights 50 emerging leaders shaping the future of AI. This recognition underscores the importance of our work to build more responsible AI, by addressing quality, safety, and compliance risks. Thank you to MAD50, and congrats to all the other innovators recognized alongside us 🙌 Learn more about the ranking 🔗 https://gisk.ar/3XygUdB #ResponsibleAI #AISafety #AISecurity

🌟 Join us today in the open-source AI developer meetup organized by Docker, Inc! 🐳🐢 Alex Combessie 🐢, our CEO, we'll demo our open-source RAG toolkit, that can evaluate RAG agents automatically. It evaluates components such as the generator, retriever, rewriter, router, and knowledge base. Don't miss demos from Hugging Face, Weaviate, Koyeb, LlamaIndex, Amazon Web Services (AWS), Red Hat, Neo4j, GaiaNet, and, Cohere. 🇫🇷 STATION F in Paris 📆 20 June at 6 PM Join us: https://gisk.ar/3VypBSo #opensource #RAG #GenAI #AIdev

We're looking for a Senior Data Scientist to lead customer projects 🎉 In this role, you'll lead red teaming AI applications for customers, identifying vulnerabilities and providing guidance to clients on mitigating risks and optimizing their LLM apps. You'll collaborate closely with our R&D, sales, and marketing teams, supporting the sales pipeline and creating technical content. If you're seeking to join a dynamic company where you can grow and help build responsible AI, this is your chance! We can’t wait to meet you 🤩 Questions or thoughts? Feel free to comment below. Apply here 👉🏻 https://gisk.ar/4esZvc3 #DataScience #LLMs #AIRedTeaming #hiring

🚀 Missed our last webinar on #RedTeaming LLMs? The replay is now available! 🎥 Follow Alexandre Landeau's talk, and learn how to enhance the security of LLM applications using red-teaming techniques. Key takeaways: - Apply cybersecurity red-teaming to LLMs - Automatically detect vulnerabilities in LLM apps - Scale your security efforts Thank you to Open Data Science Conference (ODSC) for hosting this session! 👉 Watch the replay now: https://gisk.ar/45v31in #LLMs #AIsecurity #ODSC

Follow our talk at Open Data Science Conference (ODSC)! today! 🔥 Join Alexandre Landeau for his talk, "Secure LLM App Deployments—Strategies and Tactics," where he'll explain how to enhance the security of Large Language Models (LLMs). Discover how red-teaming techniques from cybersecurity can be applied to identify and evaluate vulnerabilities in LLM applications. Learn how Giskard's tools integrate into your workflow for automatic vulnerability detection, helping you scale your security efforts for Generative AI. A big thank you to ODSC for hosting this talk 🙌 📆 June 12, 2024 ⏰ 6PM (CET) - 12PM (ET) Register now for free: https://gisk.ar/3VmTJAg #RedTeaming #LLMs #AIsecurity #ODSC

We're hiring! 🚀 We're looking for a #MLEngineer Intern to join our team! In this role, you'll participate in red-teaming LLM applications for our customers, finding vulnerabilities and adding new evaluation methods. You'll also provide technical guidance to customers on building secure and robust LLM apps. If you are passionate about responsible AI, and want to join a fun team, apply here! 🐢
 🔗 https://gisk.ar/3z6huEY #LLMs #AIRedTeaming #MachineLearning #hiring

✨ New integration with Databricks MLflow! ✨ This partnership brings together Giskard's LLM evaluation capabilities and MLflow's model management features. Databricks users can now automatically identify vulnerabilities on ML models and LLMs, generate domain-specific tests, and compare model performance across different versions. What's Giskard's open-source scan?  It ensures the automatic identification of vulnerabilities in ML models and LLMs, such as hallucinations, reliability and robustness issues. Learn more about the integration 👉 https://lnkd.in/eNxnPM_5 #databricks #mlflow #LLMeval #MLOps

How LLMOps is different from MLOps? 🪄🤔 Short for Large Language Model Operations, LLMOps focuses on the deployment, management, and optimization of LLMs in production environments. In this article, you'll discover: ✅ The key differences between MLOps and LLMOps ✅ Challenges in productionizing LLM applications (and how to overcome them) ✅ Best practices for implementing LLMOps in your organization You'll gain insights into maximizing the performance, reliability, and cost-effectiveness of your LLM deployments. Link to the article 🔗 https://gisk.ar/4c8mwzi #LLMs #LLMOps #MLOps #MachineLearning