AI Engineering Podcast

SummaryIn this episode of the AI Engineering Podcast, host Tobias Macey sits down with Ben Wilde, Head of Innovation at Georgian, to explore the transformative impact of agentic AI on business operations and the SaaS industry. From his early days working with vintage AI systems to his current focus on product strategy and innovation in AI, Ben shares his expertise on what he calls the "continuum" of agentic AI - from simple function calls to complex autonomous systems. Join them as they discuss the challenges and opportunities of integrating agentic AI into business systems, including organizational alignment, technical competence, and the need for standardization. They also dive into emerging protocols and the evolving landscape of AI-driven products and services, including usage-based pricing models and advancements in AI infrastructure and reliability.AnnouncementsHello and welcome to the AI Engineering Podcast, your guide to the fast-moving world of building scalable and maintainable AI systemsYour host is Tobias Macey and today I'm interviewing Ben Wilde about the impact of agentic AI on business operations and SaaS as we know itInterviewIntroductionHow did you get involved in machine learning?Can you start by sharing your definition of what constitutes "agentic AI"?There have been several generations of automation for business and product use cases. In your estimation, what are the substantive differences between agentic AI and e.g. RPA (Robotic Process Automation)?How do the inherent risks and operational overhead impact the calculus of whether and where to apply agentic capabilities?For teams that are aiming for agentic capabilities, what are the stepping stones along that path?Beyond the technical capacity, there are numerous elements of organizational alignment that are required to make full use of the capabilities of agentic processes. What are some of the strategic investments that are necessary to get the whole business pointed in the same direction for adopting and benefitting from AI agents?The most recent splash in the space of agentic AI is the introduction of the Model Context Protocol, and various responses to it. What do you see as the near and medium term impact of this effort on the ecosystem of AI agents and their architecture?Software products have gone through several major evolutions since the days of CD-ROMs in the 90s. The current era has largely been oriented around the model of subscription-based software delivered via browser or mobile-based UIs over the internet. How does the pending age of AI agents upend that model?What are the most interesting, innovative, or unexpected ways that you have seen agentic AI used for business and product capabilities?What are the most interesting, unexpected, or challenging lessons that you have learned while working with businesses adopting agentic AI capabilities?When is agentic AI the wrong choice?What are the ongoing developments in agentic capabilities that you are monitoring?Contact InfoEmailLinkedInParting QuestionFrom your perspective, what are the biggest gaps in tooling, technology, or training for AI systems today?Closing AnnouncementsThank you for listening! Don't forget to check out our other shows. The Data Engineering Podcast covers the latest on modern data management. Podcast.__init__ covers the Python language, its community, and the innovative ways it is being used.Visit the site to subscribe to the show, sign up for the mailing list, and read the show notes.If you've learned something or tried out a project from the show then tell us about it! Email hosts@aiengineeringpodcast.com with your story.To help other people find the show please leave a review on iTunes and tell your friends and co-workers.LinksGeorgianAgentic Platforms And ApplicationsDifferential PrivacyAgentic AILanguage ModelReasoning ModelRobotic Process AutomationOFACOpenAI Deep ResearchModel Context ProtocolGeorgian AI Adoption SurveyGoogle Agent to Agent ProtocolGraphQLTPU == Tensor Processing UnitChris LattnerCUDANeuroSymbolic AIPrologThe intro and outro music is from Hitman's Lovesong feat. Paola Graziano by The Freak Fandango Orchestra/CC BY-SA 3.0

SummaryIn this episode of the AI Engineering Podcast Kasimir Schulz, Director of Security Research at HiddenLayer, talks about the complexities and security challenges in AI and machine learning models. Kasimir explains the concept of shadow genes and shadow logic, which involve identifying common subgraphs within neural networks to understand model ancestry and potential vulnerabilities, and emphasizes the importance of understanding the attack surface in AI integrations, scanning models for security threats, and evolving awareness in AI security practices to mitigate risks in deploying AI systems.AnnouncementsHello and welcome to the AI Engineering Podcast, your guide to the fast-moving world of building scalable and maintainable AI systemsYour host is Tobias Macey and today I'm interviewing Kasimir Schulz about the relationships between the various models on the market and how that information helps with selecting and protecting models for your applicationsInterviewIntroductionHow did you get involved in machine learning?Can you start by outlining the current state of the threat landscape for ML and AI systems?What are the main areas of overlap in risk profiles between prediction/classification and generative models? (primarily from an attack surface/methodology perspective)What are the significant points of divergence?What are some of the categories of potential damages that can be created through the deployment of compromised models?How does the landscape of foundation models introduce new challenges around supply chain security for organizations building with AI?You recently published your findings on the potential to inject subgraphs into model architectures that are invisible during normal operation of the model. Along with that you wrote about the subgraphs that are shared between different classes of models. What are the key learnings that you would like to highlight from that research?What action items can organizations and engineering teams take in light of that information?Platforms like HuggingFace offer numerous variations of popular models with variations around quantization, various levels of finetuning, model distillation, etc. That is obviously a benefit to knowledge sharing and ease of access, but how does that exacerbate the potential threat in the face of backdoored models?Beyond explicit backdoors in model architectures, there are numerous attack vectors to generative models in the form of prompt injection, "jailbreaking" of system prompts, etc. How does the knowledge of model ancestry help with identifying and mitigating risks from that class of threat?A common response to that threat is the introduction of model guardrails with pre- and post-filtering of prompts and responses. How can that approach help to address the potential threat of backdoored models as well?For a malicious actor that develops one of these attacks, what is the vector for introducing the compromised model into an organization?Once that model is in use, what are the possible means by which the malicious actor can detect its presence for purposes of exploitation?What are the most interesting, innovative, or unexpected ways that you have seen the information about model ancestry used?What are the most interesting, unexpected, or challenging lessons that you have learned while working on ShadowLogic/ShadowGenes?What are some of the other means by which the operation of ML and AI systems introduce attack vectors to organizations running them?Contact InfoLinkedInParting QuestionFrom your perspective, what are the biggest gaps in tooling, technology, or training for AI systems today?Closing AnnouncementsThank you for listening! Don't forget to check out our other shows. The Data Engineering Podcast covers the latest on modern data management. Podcast.__init__ covers the Python language, its community, and the innovative ways it is being used.Visit the site to subscribe to the show, sign up for the mailing list, and read the show notes.If you've learned something or tried out a project from the show then tell us about it! Email hosts@aiengineeringpodcast.com with your story.To help other people find the show please leave a review on iTunes and tell your friends and co-workers.LinksHiddenLayerZero-Day VulnerabilityMCP Blog PostPython Pickle Object SerializationSafeTensorsDeepseekHuggingface TransformersKROP == Knowledge Return Oriented PromptingXKCD "Little Bobby Tables"OWASP Top 10 For LLMsCVE AI Systems Working GroupRefusal Vector AblationFoundation ModelShadowLogicShadowGenesBytecodeResNet == Resideual Neural NetworkYOLO == You Only Look OnceNetronBERTRoBERTAShodanCTF == Capture The FlagTitan Bedrock Image GeneratorThe intro and outro music is from Hitman's Lovesong feat. Paola Graziano by The Freak Fandango Orchestra/CC BY-SA 3.0

Julian LaNeve, CTO of Astronomer, shares his expertise on the transition from simple LLMs to complex agentic AI systems. He stresses the importance of starting with easy applications to build foundational knowledge. The discussion delves into orchestrating AI workflows using directed acyclic graphs and highlights the necessity of robust data management. Julian also addresses the challenges of reliability and observability in AI, urging teams to thoughtfully evaluate their operational readiness and investment decisions in this dynamic field.

Emmanouil (Manos) Koukoumidis, CEO of Oumi and former Google Cloud AI tech lead, talks about fostering community in AI development. He stresses the need for open-source models to promote collaboration and accessibility, likening Oumi's vision to 'the Linux of AI.' Manos shares insights on navigating the overwhelming choices in AI models and the importance of engaging a community for innovation. He also addresses gaps in AI accessibility and the need for standardization to empower both researchers and enterprises in their AI journeys.

In this engaging discussion, Adil Hafiz, co-founder of Ardenimo and an expert with a rich engineering background at Microsoft and Amazon, sheds light on the Arch Gateway. This innovative tool simplifies AI integration for developers, allowing them to focus on core functions while bypassing complex AI specifics. He highlights the project's use of Rust and Envoy to enhance performance, discusses community feedback's crucial role, and outlines future aspirations for developing a leading planning model and improving AI agent interoperability.

Ali Golshan, Co-founder and CEO of Gretel.ai, dives into the fascinating world of synthetic data and its pivotal role in advancing AI applications. He discusses how synthetic data can enhance privacy while improving the quality and structural stability of datasets. The conversation highlights the shift from traditional data methods to the use of language models and the challenges of scaling synthetic data in production. Ali also explores its transformative applications in sectors like healthcare and finance, underscoring the importance of governance and ethical considerations.

Viraj Mehta, CTO and co-founder of TensorZero, shares insights on optimizing AI applications with their innovative LLM gateways. He discusses how these gateways standardize communication and manage interactions between applications and AI models. The conversation dives into sustainable AI optimization and the challenges of integrating structured data inputs. Viraj also highlights the role of user feedback in enhancing AI interactions, as well as the architectural innovations that improve efficiency and usability for developers.

Ron Green, co-founder and CTO of Kung Fu AI, dives into the evolving AI landscape and the complexities of generative AI engines. He discusses the limitations of large language models and the critical need for human oversight and robust data management. Ron highlights innovative methods like Retrieval-Augmented Generation and the significance of targeted, domain-specific AI solutions. He expresses optimism for AI's future, predicting major advancements in the next 20 years that integrate seamlessly into everyday applications.

Jim Olson, CTO of ModelOp, specializes in generative AI governance and regulations. He discusses the importance of monitoring and inventory for compliance in high-risk areas like healthcare. Olson emphasizes the need for technical controls to manage data governance and the continuous monitoring of AI models to detect issues. He addresses the balance between innovation and regulation, particularly in light of evolving EU regulations, and highlights the necessity of building trust through effective governance solutions.

Vasilije Markovich, a data engineer and AI specialist from Montenegro, discusses enhancing large language models with memory. He highlights the challenges of context window limitations and forgetting in LLMs, introducing hierarchical memory to improve performance. Vasilije dives into his creation, Cognee, which manages semantic memory, emphasizing its potential applications and the blend of cognitive science with data engineering. He shares insights from building an AI startup, the importance of user feedback, and future developments in open-source AI technology.