Loading…
Browse by type
Explainer reads
Explainer posts on Yomu, summarised newest first.
- 1,369 posts
- 20 companies
The Good Stuff
Paid picks we use ourselves. Links carry a referral code — a subscription through one earns us a commission at no extra cost to you.
Explainer posts
How DOCA GPUNetIO Unifies GPU-Initiated Networking Across the NVIDIA Software Stack
NVIDIA DOCA GPUNetIO is a GPU-centric networking SDK that enables CUDA kernels to drive Ethernet, RDMA, Verbs, and DMA operations with the CPU removed from the application critical path. It unifies previously fragmented GDA-KI-style RDMA implementations from libraries such as NCCL, NVSHMEM, UCX, and the Aerial 5G SDK into a single shared foundation. The framework provides both a full DOCA SDK version and a lighter open-source Verbs-focused implementation; the open-source version can optionally dlopen the SDK at runtime. On the data path, CUDA kernels directly post Work Queue Entries, ring the NIC doorbell, and poll for completions without CPU intervention. Benchmarks using NVSHMEM show that the GPU-initiated path avoids the CPU proxy bottleneck, achieving higher bandwidth for small message sizes compared to CPU-mediated communication.
Control How Your GPU Shares Work with Green Contexts
Green contexts, available in the CUDA Driver API since 12.4 and the Runtime API since 13.1, let applications explicitly partition GPU execution resources such as SMs and workqueues. They are lightweight to create and destroy, do not implicitly synchronize unrelated work, and allow provisioning workqueues to reduce false dependencies. Streams are created from a green context rather than relying on implicit device state, making targeting explicit. Benchmarks on a Blackwell GPU show a critical kernel completing in 0.007 ms with green context partitioning, compared to 0.140 ms with high-priority streams and 3.727 ms with equal priority. The tradeoff is fewer SMs for the bulk workload, but green contexts are opt-in and can be adopted incrementally alongside existing CUDA code.
Rossoctl for agent discoverability, security, and observability on Kubernetes
Agent sprawl arises when many AI agents on Red Hat OpenShift become redundant, insecure, and untraceable, using static API keys over HTTP and lacking visibility into reasoning failures. Rossoctl, a Kubernetes-native controller, automates discovery by declaring AgentRuntime resources that fetch metadata to create AgentCard custom resources, and uses a two-phase Helm deployment to resolve the URL dependency between routes and AgentCards. It adopts the A2A protocol, which standardizes agent interaction through a machine-readable AgentCard at /.well-known/agent-card.json and a JSON-RPC interface for consistent payloads. Security follows a six-layer zero-trust model: SPIRE issues short-lived certificates, Istio enforces mTLS, AgentCards are signed with JWS, and AuthBridge provides scoped OAuth2 tokens. Observability uses opt-in MLflow tracing that captures orchestration flow, tool calls, and model latency, enabling diagnosis of semantic failures such as hallucinations or data leaks that standard metrics miss.
Our Framework for Reviewing AI-Generated Code
JetBrains Research, in collaboration with Lund University, studied how to review AI-generated code as AI agents produce thousands of lines across multiple files quickly. The core problem identified is trust calibration: developers cannot gauge an LLM’s confidence per line, making line-by-line review unscalable. The study proposes a three-level conceptual framework: overview for high-level hypotheses, file-level risk stratification, and code snippet-level chunk decomposition with chain-of-thought linkage. This framework was developed through participatory design with 17 practitioners and a follow-up survey of 43 professionals, and will be presented at ESEIW 2026. The paper argues that current diff viewers are inadequate for AI-generated code and that tools must surface risk and confidence signals at the granularity where developers allocate attention.
Content and Community Basics for JetBrains Content Creators
The post covers community-building basics for JetBrains Content Creators in a golden age of digital creation with high demand and AI adoption, accepting new creators daily. Creators should include the JetBrains promotional discount code in video descriptions, with an example offering 20% off for new and existing subscriptions, and they can explore all JetBrains IDEs through the All Products Pack to diversify their workflow. Maintaining active social media is essential for distribution, and the program supplies news-worthy items like merch drops and IDE updates for creators to share with their audience. Staying in touch through communication, challenges, and giveaways is crucial, and the program provides up to twelve free IDE subscriptions per creator for giveaways. Even a simple subscription plus repost formula works effectively, and all new creators are welcome to join the program.
Data Modeling and Schema Design for MongoDB Search Part 1: Practical Considerations
Data modeling for MongoDB Search requires balancing conceptual, logical, and physical schema decisions against workload demands and system constraints. The $search stage must be the first in an aggregation pipeline, queries a single Search index, and runs against a single collection, with results returned in relevancy order. The underlying Lucene engine supports two mappings for arrays of sub-documents: default document mapping flattens fields into one index object, while embeddedDocuments mapping creates separate Lucene index objects per sub-document. Embedded mapping enables cross-field matching within sub-documents but explodes the index object count and causes heavy reindexing on any update, as the entire block is deleted and re-added. This post establishes the foundational concepts and physical constraints that will inform schema patterns in Part 2.
Data Modeling and Schema Design for MongoDB Search Part 2: Use Cases and Patterns
The post presents best practices for data modeling and schema design in MongoDB Search, covering use cases and patterns such as the Entity Pattern to guide developers in structuring collections for effective search. It emphasizes that searchable items, like movies, cast members, or genres, should be modeled as first-class documents in their own collections rather than as nested structures or ambiguous attributes. The Entity Pattern addresses ‘find anything anywhere’ and as-you-type suggestions by representing all domain entities as documents in a specialized collection with stable identifiers and a type field for filtering. The Attribute Pattern is explicitly discouraged for search because its key-value structure prevents type-sensitive field mapping, making it better to model attributes as straightforward fields. The post details a multi-indexing strategy on the name field that applies left-edge, phonetic, shingled, and language-agnostic analysis to enable flexible matching and relevancy scoring, with optional type-specific metadata nested in sub-documents.
Data-Driven, Interoperable Core Banking: A BIAN and MongoDB Perspective
Core banking modernization is hindered by semantic fragmentation, where the same banking concepts are defined differently across systems, creating duplicated logic and fragile integrations. The Banking Industry Architecture Network (BIAN) provides a shared semantic foundation by defining common banking domains, boundaries, and service interactions, enabling a composable architecture. MongoDB’s document model, using JSON as both storage and interchange format, allows banks to represent business domains flexibly and evolve incrementally without schema migrations, eliminating the ORM impedance mismatch. MongoDB ensures financial integrity through single-document atomicity for common ledger updates and multi-document ACID transactions for complex cross-domain workflows, while Change Streams enable event-driven operations. Together, BIAN and MongoDB create an operating model for continuous transformation, where teams modernize domain by domain, supported by a shared business language and a scalable, flexible data platform.
NEAREST BY Join: Scaling Vector Search in Databricks Runtime
Batch vector search workloads, such as entity resolution and semantic tagging, require high throughput across millions of queries against billions of vectors, unlike real-time serving. Databricks built NEAREST BY, a first-class top-k ranking join syntax, directly into its runtime engine rather than federating to an external endpoint. The feature is implemented as a logical join that the optimizer lowers onto standard relational operators, enabling fault tolerance and elastic parallelism. A fused Photon operator collapses the plan’s middle section with a blocked GEMM kernel designed to approach peak arithmetic intensity for vector scoring. An optional IVF index, stored as an ordinary liquid-clustered Delta table, allows approximate queries to probe only relevant clusters while preserving the same primitives and correctness guarantees.
AI Agents Need Memory: Building Persistent Memory for Agentforce
Enterprise AI agents without persistent memory suffer from the “Groundhog Day” problem, repeatedly asking users to re-explain themselves. Agentforce Memory addresses this by giving agents persistent memory that retains user preferences, conversation history, and context across sessions. The team first established a shared terminology document to define concepts like semantic and episodic memory before architectural discussions could proceed. The architecture evolved incrementally, starting with conversation history injection and later adding semantic retrieval to find relevant past information regardless of recency. Memory retrieval uses keyword search during pre-orchestration to balance context quality with low latency, and moving memory into the Reasoner simplified configuration for administrators. Privacy and user trust were treated as architectural requirements from the start, enabling individual user controls over memory creation and management.
Intercom on Product: How we became an AI-first company - The Intercom Blog
Intercom’s Chief Product Officer and Co-founder discuss how generative AI reshaped their strategy, product development, and internal rituals. They describe moving from isolated AI tasks to AI-native products built from the ground up with AI at their core. A key move was launching Fin, their AI agent, as a standalone product on competitor platforms like Zendesk, prioritizing AI leadership over platform exclusivity. They draw parallels to Facebook’s mobile pivot and the iPhone’s initial skepticism, arguing generative AI will face similar resistance but have profound disruptive potential. The conversation emphasizes that second- and third-order effects of AI will create entirely new workflows and behaviors, making full impact hard to predict.
Start Good Trouble: Mastering SaaS metrics with Equals' Bobby Pinero - The Intercom Blog
Equals co-founder and CEO Bobby Pinero joined the Start Good Trouble webinar to discuss how SaaS metrics unlock business potential and tell a compelling story. He argued that spreadsheets remain central despite the explosion of new BI tools, because everything eventually ends up back in a spreadsheet connected to live data. Drawing on seven years as Intercom’s senior director of finance, where he scaled ARR from under $1 million to over $150 million, Pinero highlighted the complexity of early-stage metrics like CAC, LTV, and retention. Equals recently published The Guide to SaaS Metrics to help leaders understand, optimize, and report on key drivers, and Pinero emphasized the need to tailor metrics to the business model and start with the story. He also recommended writing consistently and authentically to clarify ideas and connect with audiences, and noted that horizontal products require narrowing down use cases to scale effectively.
This is getting real: Welcome to the AI revolution - The Intercom Blog
The talk traces the tech industry’s journey from the late stages of SaaS in 2019, through pandemic virtual offices and NFTs, to the 2022 downturn and ChatGPT’s launch in November 2022. 2023 was a year of prototypes and ‘thin wrappers’ on ChatGPT, but also real learning about building with LLMs, including non-deterministic approaches. The author argues the AI transformation is now real, moving beyond superficial features to fundamentally rethinking products, exemplified by the shift from ‘Software as a Service’ to ‘Service as Software.’ The future of customer service includes proactive, personalized bots, video and voice bots, and the ability to analyze all conversations to deeply understand customers. The author cites Bloom’s 2 sigma effect as an example of AI’s potential in education and urges optimism, noting that learners will inherit the earth.
The Ticket: Stacy Sherman on how to design customer experiences that drive loyalty - The Intercom Blog
Customer expectations have never been higher, and the experience a business delivers can determine whether it wins or loses customers. Stacy Sherman, a CX author and speaker, joined The Ticket podcast to discuss her Doing CX Right framework for designing standout customer experiences. She emphasizes that companies must move from merely planning to taking action, and that walking in customers’ shoes to get the basics right is essential. Sherman advises validating every approach with real customers and ensuring internal teams feel valued to create a culture of customer excellence. She also recommends mapping the customer journey to identify where AI can enhance interactions and where the human touch should be preserved.
Intercom on Product: The evolution of product management in the age of AI - The Intercom Blog
Product management is at a critical juncture as generative AI reshapes how products are conceived, built, and used. Indiscriminately adding AI is risky; product managers must ensure the technology solves real user problems rather than being an end in itself. Curiosity is essential for driving exploration of generative AI’s new possibilities. A foundational understanding of AI is important, but the ability to collaborate with AI and ML experts is even more critical for bridging technical possibilities with customer needs. The future of product management will be shaped by those willing to learn, adapt, and explore the art of the possible.
How AI is Transforming Customer Service Tech Stacks
Customer service tech stacks have become fragmented and outdated, comprising many disconnected point solutions that slow teams and prevent fast, personal experiences. Recent advances in AI now enable a consolidated AI-first platform with AI at the core, empowering teams to analyze conversation data, spot trends, and deliver personalized support at scale using business and customer data. AI bots can first resolve repetitive queries 24/7, then build contextual experiences from historical data, and eventually take actions like setting up capabilities for customers. AI makes proactive engagement possible by detecting customer actions to preemptively unblock pain points, onboard users, and prevent friction. This shift improves metrics like customer satisfaction and operational efficiency and supports data-driven programs like Intercom’s Beyond CSAT to measure success.
The Ticket: Amplitude’s Jenna Elliott on navigating the future of customer success - The Intercom Blog
On The Ticket podcast, Intercom’s Declan Ivory interviews Amplitude’s Jenna Elliott about navigating AI-driven customer experience. Elliott argues a data-driven culture requires both qualitative and quantitative data to gauge team efficacy and identify areas for improvement, enabling targeted strategies that drive quicker value for customers. She highlights the balance between operational excellence and customer centricity, requiring a focus on KPIs and OKR frameworks as well as agile responsiveness to customer needs. AI should be implemented thoughtfully to complement rather than replace human interaction, boosting productivity and creating real value when integrated intentionally into the customer journey. New roles such as technical writers, customer journey designers, and strategists will emerge to ensure data is used in the right way to inform the customer journey.
Understanding the Economics of Customer Service in the Age of AI
Customer service has long been perceived as a cost center rather than a value driver, but the transformative arrival of generative AI, including tools such as Fin, is rapidly changing that equation. Understanding the economics of customer service involves balancing the financial costs of providing support against the value it creates, aiming for sustainability and alignment with overall business goals. AI automates routine tasks, enabling support teams to focus on more complex and value-adding activities that require stronger troubleshooting skills, subject matter expertise, and customer empathy. Integrating AI requires investment beyond just purchasing the technology, including knowledge management, content creation, and building an ecosystem that supports continuous learning by support teams. This alters the traditional linear scaling model, enabling companies to absorb increased workload without proportional expansion, leading to cost savings and more rewarding roles.
The Ticket: Simon T. Bailey on the platinum service principles that create lifelong customers - The Intercom Blog
The blog post recounts a podcast episode from The Ticket featuring Simon T. Bailey, author of Be the Spark, who discusses platinum service principles to create lifelong customers with Intercom VP of Customer Support Declan Ivory. Bailey’s pivotal realization came while sweeping streets at Disney, where he understood his true purpose was to create magical moments for families seeking a once-in-a-lifetime experience. He argues the platinum standard for customer service supersedes the gold standard by emphasizing personal connection, engagement, and making people feel valued rather than merely completing transactions. Key enablers include employee empowerment through recognition and sharing success stories, and leadership that avoids pitfalls like micromanagement while modeling the behaviors that enhance experience. Cultivating resilience and a genuine customer-love mindset drives loyalty, advocacy, and positive business outcomes, enabling teams to maintain high-quality service despite challenges.
Intercom on Product: How to design standout products in an AI world - The Intercom Blog
Intercom’s VP of Design Emmet Connolly and Staff Product Designer Molly Mahar discuss how product designers must adapt their methods for AI, drawing from their experiences building Intercom’s products. They explain that designing for AI means moving from deterministic to probabilistic systems, where unpredictability is the norm and designers must design flexibly around surprises. Data quality is foundational for reliable AI, so designers should prioritize it; realistic prototyping using actual data helps capture essential user feedback and evaluate output effectiveness. Extensive testing is necessary to anticipate possible system failures; if live versions catch designers off guard, testing must be more robust. Building user trust requires consistent positive experiences and transparency about how the system works and its data, enabling designers to create AI experiences that make users feel comfortable and confident.
Top topics
- Architecture833
- Performance814
- LLMs659
- Machine Learning621
- Open Source563
- AI Agents551
- AI467
- Developer Experience448
Weekly digest
The best engineering reads, once a week.