The InfoQ Podcast

Shuman Ghosemajumder, Founder and CEO of Reckon and former head of Google’s Trust & Safety group, dives into the challenges of preserving digital trust. He highlights the alarming rise of deepfakes and automated social engineering in cybercrime, stressing the need for enhanced defenses. Shuman suggests leveraging behavioral telemetry and game theory to outsmart attackers. He also discusses the risks posed by generative AI in fraud, advocating for a zero-trust approach and smarter resource allocation in security strategies.

David Blank-Edelman, a leading figure in site reliability engineering (SRE) with nearly 40 years of operations experience, dives into the intricate relationship between software architecture and SRE. He challenges the conventional idea of seeking root causes for failures, emphasizing instead the importance of understanding what works in a system. The discussion highlights designing for reliability, embracing emergent properties, and learning from successes as pivotal to improving system resilience and collaboration between architects and SREs.

Vivek Yadav, an engineering manager at Stripe, shares his expertise in crafting a multi-year regression testing framework using Apache Spark. He highlights the importance of testing migrations against extensive historical data to avoid user regressions. Spark's parallel processing capabilities allow efficient bulk request replays. Vivek discusses the design of reusable libraries and controlled testing environments, boosting developer confidence while maintaining low costs compared to traditional database methods. He emphasizes the framework's versatility for what-if analyses and projections.

Marina Moore, a research scientist at Adara and co-chair of the CNCF Security & Compliance TAG, dives into the vulnerabilities of container security. She discusses the risks tied to sharing the host kernel and advocates for micro-VMs as a solution to reduce attack surfaces. Marina highlights the dangers of GPU inference in multi-tenant environments and critiques current isolation techniques. She also emphasizes the potential of using memory-safe languages like Rust to enhance security in cloud architectures.

In a compelling discussion, Randy Shoup, a seasoned distributed-systems architect with experience at eBay and Google, delves into the importance of learning from software failures. He advocates for blameless postmortems to foster culture and resilience. Randy shares practical strategies, like a five-step postmortem framework for understanding outages. He emphasizes modeling real-world asynchronous systems through workflows and events for better reliability, and discusses how shared trauma can enhance team cohesion in the wake of failures.

Magdalena Picariello, an AI practitioner and academic, emphasizes making AI development more business-impactful. She discusses the need to shift from traditional metrics to evaluating real-world business outcomes. Magdalena talks about implementing iterative testing systems for generative AI and prioritizing high-value edge cases. She shares insights on a data-driven, test-first approach, the importance of human-crafted tests, and tools for effective evaluation. Lastly, she highlights translating business KPIs into code to ensure alignment with user needs.

Jenish Shah, a backend engineer at Netflix with expertise in distributed systems, shares insights on effective error handling in complex environments. He discusses the evolution of microservices and the significance of clear error messages for user experience. Jenish introduces his centralized exception library, designed for uniformity across various protocols like HTTP and gRPC. He emphasizes the importance of observability metrics in error detection and offers advice on building reusable solutions to enhance engineering consistency.

Shweta Vohra, a lead architect at Booking.com and cloud-native expert, joins Stefian Weas, InfoQ lead editor, and Matt Saunders, head of DevOps at Adaptavist, to discuss pivotal trends in cloud and DevOps. They explore the rapid AI adoption in cloud services and the challenges of integrating legacy systems. Shweta emphasizes the rise of FinOps and Kubernetes' dominance, while Matt reflects on the growing executive interest in developer platforms. The panel predicts a shift towards maturing digital sovereignty and the significance of optimizing cloud costs amidst AI budgeting pressures.

Nimisha Asthagiri, Global Director in Data and AI at ThoughtWorks, chats about the significance of systems thinking and mental models in architecture. She reveals how misalignments in technology can lead to societal issues, emphasizing responsible AI practices. Nimisha discusses scaling multi-agent systems and the delicate balance of human involvement in tech operations. She envisions a harmonious relationship with AI that respects human agency, and shares her passion for simplifying complex designs while advocating for better architectural outcomes.

Elena Samuylova, Co-founder and CEO of Evidently AI, shares her expertise in LLM-powered application evaluation. She highlights the importance of distinguishing between model and system evaluations. Elena introduces the concept of using an LLM as a judge, discussing its benefits and limitations. She emphasizes the workflow for LLM evaluation, including iterative checks and stress testing. Furthermore, she advises on designing custom LLM judges and stresses the significance of team roles in this process, encouraging developers to adapt their skills as the field evolves.