From InfoQ · 40 stories
Vercel Labs Releases Zero, a Graph-First Systems Language for AI Agents
Vercel Labs introduced Zero, an experimental systems programming language designed for AI agents to read and repair compiler output. This language emphasizes a graph-first authoring workflow and provides structured error diagnostics with machine-readable repair plans, shifting the focus from human developers to AI agents for code interaction.
eslint-rspack-plugin 5.0.0 Released as Pure ESM Package
eslint-rspack-plugin version 5.0.0 has been released as a pure ESM package, removing its CommonJS build to align with the Rspack ecosystem. This change mirrors Rspack 2.0's move to pure ESM, aiming for consistent module loading practices.
HubSpot Redesigns JITA Authorization with Rule Engine for Improved Observability
HubSpot has re-engineered its Just-In-Time Access (JITA) authorization system to use a rule engine architecture, replacing complex conditional logic. This change allows for greater observability and explainability of access decisions, making it easier for engineers to understand why requests are approved or denied and to manage evolving authorization requirements.
Jotai v2.20 Released with Reworked Internal Store Building Blocks for Performance
Jotai, a React state management library, released version 2.20.0, which focuses on improving performance in high-throughput scenarios by reworking its internal store building blocks. This update addresses a performance regression introduced in earlier versions and prepares the library for a future v3 release.
Expedia Develops AI-Assisted Service Telemetry Analyzer (STAR) for Incident Investigation
Expedia Group has launched Service Telemetry Analyzer (STAR), an internal AI-assisted observability platform designed to help engineers investigate production incidents by analyzing service telemetry and generating root cause assessments. This system aims to reduce the time engineers spend identifying the source of service degradation by combining operational metrics with large language models through predefined diagnostic workflows.
QCon AI New York 2026 Registration Opens for December Conference
Registration has opened for QCon AI New York 2026, a two-day conference scheduled for December 15-16, 2026, in Jersey City, New York. The event targets senior software engineers, architects, and engineering leaders who manage AI systems in production, focusing on practical challenges encountered when deploying AI features.
Luca Mezzalira Discusses Software Architecture as a Socio-Technical Discipline
Michael Stiefel and Luca Mezzalira discussed software architecture, emphasizing its socio-technical nature and the need to adapt architectural choices to an organization's culture. They highlighted a shift towards architects acting as coaches, empowering developers through guardrails and fitness functions, and the role of collaborative learning in architectural education.
InfoQ Podcast Discusses 2026 AI, ML, and Data Engineering Trends
An InfoQ podcast featuring its AI and ML editorial staff and guests discussed current trends in AI, ML, and data engineering for 2026. The discussion covered shifts from large to small language models, the rise of AI agents, and developments in physical AI. Key predictions include increased adoption of System One models and more innovations in US-based robotics.
The Reinvention of the Dev Team: From Velocity to Agentic Coding
The software development industry has evolved from slow, infrequent releases to continuous delivery, driven by methodologies like Agile and DevOps. The focus is now shifting from maximizing velocity to understanding and implementing agentic coding, where AI agents generate code from specifications.
QCon London 2026 Security Track Host Discusses Future Cybersecurity Trends
Chris Swan, security track host at QCon London, discussed emerging cybersecurity trends, including hardware memory safety, automated governance, and post-quantum cryptography. The discussion highlighted a shift towards automated, system-level defenses and the increasing importance of supply chain security.
DoorDash GenAI Platform Team Shares Development Journey and Operating Principles
DoorDash's GenAI Platform team detailed its journey in building a generative AI platform, outlining the principles, strategic bets, and adaptations made since its inception in April 2023. The team focused on supporting product engineers with end-to-end workflows, evolving from initial model integration to developing workflows and agents.
Software Engineering Skill Development Requires Slowing Down, Says Chelsea Troy
Chelsea Troy presented at Craft conference, stating that software engineering skill development requires slowing down, despite the availability of tools and AI that accelerate outcomes. She argues that engineers must understand how systems work to learn effectively, and that senior developers can coach juniors using specific learning techniques. The core idea is that inefficient, focused engagement with difficult concepts is necessary for building cognitive capacity.
Bank Builds Cloud-Native Platform Culture Through Collaboration and Open Standards
Marcy Paramonova and Stéphane Cusin discussed building a collaborative platform culture in a bank at KubeCon & CloudNativeCon Europe. They emphasized that a platform is a collaboration system, not just infrastructure, and highlighted the importance of shared standards and an engineering mindset for cloud-native transformation.
Simon Rohrer Proposes Value Center Thinking Over Autonomous Teams in Software Development
Simon Rohrer presented a concept of "value center" thinking at the Craft conference, suggesting a shift from product-focused autonomous teams to a value-driven organizational model. This approach redefines team purpose based on delivered value and acknowledges complex dependencies, moving beyond traditional autonomous team structures.
Artie Shevchenko Discusses AI's Role in Code Reviews and Maintaining Intellectual Control
Artie Shevchenko presented at Craft conference on integrating AI into code reviews, suggesting mandatory AI checks with manual spikes for complex changes. He argued that while AI can increase velocity by handling low-risk reviews, human oversight is crucial to prevent cognitive debt and maintain intellectual control over codebases.
Scott Jenson Discusses Desktop OS Stagnation, Local-First Software, and Agentic UX
UX designer Scott Jenson discussed the lack of innovation in desktop operating systems, the limitations of mobile and cloud-centric models, and the potential of local-first software and agentic user experiences. He proposed that future operating systems should move beyond current interface design to create more intuitive and privacy-conscious computing environments, especially with the integration of large language models.
Netflix's Commerce Architecture Evolution from DVD to Global Streaming
Netflix's principal engineer Kasia Trapszo detailed the evolution of the company's commerce platform, which was originally designed for a DVD mail business but adapted to support its global streaming expansion in 2016. The platform's transformation involved significant engineering challenges and trade-offs to scale for international payment processing and user sign-ups. This evolution highlights the complexities of scaling legacy systems to meet rapid business growth.
eBPF and AI Integration for APIs Discussed by Isovalent Engineer
Dan Finneran, an engineer at Isovalent (now part of Cisco), discussed the integration of eBPF with AI and APIs, highlighting the increasing use of AI-generated code in production environments. He noted a growing disconnect where developers submit AI-generated code without understanding or being able to support it, creating maintenance challenges.
Extending Reactivity to the Server for Local-First Application Development
A presentation argues that the next significant advancement in web development lies in extending client-side reactivity to the server, moving away from traditional data fetching. This approach, termed "sync engine architecture," enables instant UI updates through optimistic state and background synchronization, improving user experience.
Challenges of Using LLMs for Selection Systems in Procurement Automation
Jendrik Jordening discussed the difficulties of integrating Large Language Models (LLMs) into procurement automation for selection tasks, highlighting issues like context window limitations, integration with existing infrastructure, and new security vulnerabilities. The presentation outlined how LLMs, despite their apparent flexibility, introduce complexities in data handling, accuracy measurement, and system reliability compared to traditional deep learning models.
Increased Incident Counts Can Signal Improved Incident Management, Not Declining Reliability
A recent article argues that a rising number of reported incidents often indicates an improvement in an organization's incident management culture, rather than a decline in system reliability. As teams enhance processes and tooling, they become more willing to formally declare incidents, leading to greater visibility into operational issues. This perspective challenges the use of incident volume as a primary performance indicator for system reliability.
Podcast Discusses Early-Stage Startup Engineering Team Challenges and Career Choices
A podcast featuring David Gudeman discussed the unique culture of early-stage startup engineering, highlighting how founder personalities and premature process implementation can negatively impact teams. The discussion also covered how engineers can build influence and make deliberate career choices within this environment.
CHERI: A Capability-Hardware Enhanced RISC Instruction Set Architecture for Memory Safety
CHERI is a hardware-based approach to memory safety and fine-grained compartmentalization, not a standalone ISA but an extension to existing ISAs like ARM's Morello and RISC-V's RVY. This technology addresses the challenge of secure communication between isolated workloads, which traditional MMU-based isolation struggles with. CHERI matters because it offers a hardware-level solution to memory safety, potentially reducing vulnerabilities in systems from mobile phones to data centers.
KubeCon Speakers Discuss Shifting from Project to Product Thinking for Platform Development
Eugenia Bergman and Hagen Tonnies presented at KubeCon & CloudNativeCon Europe on their shift from project-oriented to product-oriented thinking for platform development, emphasizing user adoption and clear interfaces. This approach aims to reduce coordination overhead and align development with actual user value by treating internal platform capabilities as products with producers and consumers.
JioHotstar Details Distributed Engineering for Personalized Ad Requests at Scale
JioHotstar published an engineering overview of its ad request workflow, explaining how it coordinates distributed services to select and deliver personalized advertisements during video playback. This matters as it details how a large streaming platform handles real-time ad decisions under strict latency and high traffic conditions.
Disaggregated Systems Driven by Cloud Economics and Network Advances
Murat Demirbas, from MongoDB Research and formerly AWS, discussed the rise of disaggregated systems, emphasizing that cloud economics and advancements in high-speed networking are the primary drivers. This architectural shift decouples compute and storage to address their inherent mismatches and optimize costs for cloud providers and users.
DORA 2025 Research Identifies Team Profiles and Capabilities for AI-Assisted Software Development
DORA's 2025 research on AI in software development identifies seven team profiles and seven capabilities that enhance AI adoption outcomes. The research emphasizes that AI acts as an amplifier, with strategic focus on organizational systems yielding the greatest returns.
Johannes Schickling Discusses Transition to Client-Side Event Sourcing and Local-First Development
Johannes Schickling, founder of Prisma, discussed his shift from traditional three-tier web stacks to a local-first approach using client-side event sourcing and SQLite for his music curation app, Overtone. This transition highlights the architectural considerations and trade-offs involved in moving data bottlenecks from servers to clients for improved performance and responsiveness.
Platform Team Improves Compliance Adoption by Simplifying Governance and Prioritizing Developer Experience
A platform team improved compliance adoption by simplifying governance, prioritizing key issues, and rolling out changes incrementally, as detailed by Davide de Paolis. This approach addressed initial developer friction caused by new workflows and poor documentation, leading to successful adoption through prevention, detection, and communication.
DBOS Project Proposes Database-Centric Architecture for Workflow Orchestration
A presentation introduced the DBOS project, an architectural approach that compiles application workflows directly into existing databases. This method aims to simplify the management of distributed system state and improve failure handling, particularly for complex AI-driven processes.
Using stacked pull requests to manage large AI-generated code contributions
This article explains how to use stacked pull requests to manage large code contributions, particularly those generated by AI agents. It addresses the challenge of reviewing extensive, single pull requests that AI agents often produce, which can lead to under-reviewed code and conflict-prone processes.
AI Tools Challenge the Relevance of Traditional LeetCode-Style Coding Interviews
A developer argues that LeetCode-style coding interviews are outdated, especially with the advent of AI tools that can generate code. The author recounts failing a basic coding interview question, highlighting the disconnect between interview practices and real-world development aided by AI.
Finding Resilience Bugs in Systems Before They Are Built
Hillel Wayne discusses a method for identifying resilience bugs in system designs before development begins. This approach involves analyzing system descriptions to proactively uncover potential failures and race conditions, contrasting with traditional agile methods that prioritize testing completed code.
Form3 Details Multi-Cloud Strategy for Financial Services to Address Regulatory Concerns
Form3, a payment processing company, outlined its approach to running its platform on three different cloud providers simultaneously. This strategy was developed in response to UK banking regulator concerns about concentration risks associated with financial institutions relying on a limited number of cloud providers.
Integrating Rust Incrementally into Existing Codebases for Performance
Lily Mara, a staff engineer at Discord, presented on how to improve software performance by gradually adding Rust code to existing projects. This approach avoids the common pitfalls and complexities associated with full software rewrites.
IOP Systems Co-founder Discusses eBPF-based Telemetry and Low-Overhead Metrics
Brian Martin, co-founder of IOP Systems, presented on instrumentation at scale, focusing on the open-source eBPF telemetry agent Rezolus and the low-overhead metrics library metriken. The presentation details how these tools provide fine-grained system visibility for performance-critical applications.
Jochen Joswig Discusses Sustainable Testing Strategies at OOP Conference
Jochen Joswig presented strategies for sustainable testing at the OOP Conference, focusing on reducing unnecessary test executions and optimizing resource use. These methods aim to decrease testing time and energy consumption by running only essential tests and those affected by code changes.
Using Numba to Optimize Python for Financial Models
A presentation discusses how Numba can accelerate Python code for computationally intensive financial modeling in the life insurance industry. This approach allows organizations to leverage existing Python resources for high-performance tasks, addressing Python's inherent slowness in numerical operations.
Continuous Delivery for Foundational Platforms: Challenges and Approaches
Ian Nowland, co-founder of Junction Labs and former AWS EC2 engineer, presented on the challenges of applying continuous delivery (CD) practices to foundational platforms. The presentation distinguishes between foundational and non-foundational platforms, highlighting why standard CD models are often insufficient for critical infrastructure.
Understanding Progressive Collapse: Applying Civil Engineering Lessons to Digital Systems
A presentation explores the concept of progressive collapse, originating from civil engineering, and applies it as a metaphor to understand and prevent cascading failures in distributed digital systems. The discussion uses the 1968 Ronan Point tower block collapse as a primary example to illustrate the concept.