# knowledge-graph Published articles for knowledge-graph. This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## Unified Knowledge Graph RAG on AWS: GraphRAG and LightRAG on one stack DevFeed: [Unified Knowledge Graph RAG on AWS: GraphRAG and LightRAG on one stack]() Original publisher: [Read original article]() Author: Jonas Kim Published: 2026-09-14T16:55:58Z Content type: article Language: en Sources: [AWS Open Source Blog]() Topics: [Retrieval Augmented Generation (RAG)](), [Graphs](), [Amazon Bedrock](), [Amazon OpenSearch Service](), [Amazon Web Services]() Tags: [amazon-bedrock](), [amazon-opensearch-service](), [aws](), [knowledge-graph](), [open-source](), [rag](), [retrieval](), [search](), [vector-search]() ### AI overview This article presents a unified knowledge-graph RAG stack that brings Microsoft GraphRAG and HKUDS LightRAG together on Amazon Bedrock, Amazon Neptune, and Amazon OpenSearch Service. The shared stack supports common ingestion, indexing, caching, and multilingual handling while allowing the retrieval methodology to be selected per query. ### Source excerpt Picture a compliance analyst staring at a few thousand contracts, amendments, and internal memos, trying to answer one question that sounds straightforward: "Which of our obligations are exposed if this one milestone slips?" The answer isn't written in any single document. It's stitched across three -- a master agreement that ties a payment to a [...] ## How to measure and improve instrumentation quality for better full-stack observability DevFeed: [How to measure and improve instrumentation quality for better full-stack observability]() Original publisher: [Read original article]() Author: Arpit kumar Published: 2026-09-12T11:22:06.456390Z Content type: article Language: en Sources: [Grafana Labs blog on Grafana Labs]() Topics: [Instrumentation](), [observability](), [Grafana Cloud](), [telemetry]() Tags: [analytics](), [grafana](), [grafana-cloud](), [how-to](), [instrumentation](), [knowledge-graph](), [logs](), [metrics](), [monitoring](), [observability](), [traces]() ### AI overview This article explains how to measure and improve instrumentation quality for full-stack observability. It introduces Grafana Cloud's continuously updated instrumentation quality report, which evaluates each service using automated checks for telemetry completeness, correctness, signal coverage, naming, Kubernetes correlation, and connections across the service graph. ### Source excerpt Modern engineering teams instrument everything, with metrics, logs, traces, and profiles flowing from hundreds of services at once. But full-stack observability isn't really about collecting more telemetry; it's about having a single, unified picture of how your services connect to every layer beneath them, including their dependencies, the pods and nodes they run on, and the logs, traces, and profiles that explain their behavior. But there's often a quiet problem hiding underneath all that data: not all instrumentation is created equal, and every gap silently breaks one of those connections. One service, for example, might emit metrics but no logs, so when you pivot from "this is erroring" to "show me why," you hit a dead end. Another might have logs but an invalid service.name or a missing k8s.pod.name that breaks correlation, dropping it out of the graph and away from its pods and nodes. A third service might look perfectly healthy right up until an incident, when you discover its traces were never wired up and the trail goes cold exactly when you need it most. To fix this, Grafana Cloud's Knowledge Graph now includes an instrumentation quality report: an automated, continuously updated assessment of how well each of your services is instrumented--and, in effect, how they plug into the full-stack picture. In this post, we'll walk through how to read the instrumentation quality report, how the scoring works, and how to use it to systematically raise the observability bar, keeping every layer of your stack joined up across every service you run. What is instrumentation quality? Instrumentation quality is a measure of how complete and correct the telemetry for a given service is, judged against a set of automated checks. Each service is evaluated by a server-computed set of quality checks: small, focused rules that validate one specific thing about a service's telemetry. A few examples: Does the service emit logs? Are service graph metrics present? Is the service nam ## Knowledge Graph as context for LLMs: demonstrating decisive RCA and faster production performance DevFeed: [Knowledge Graph as context for LLMs: demonstrating decisive RCA and faster production performance]() Original publisher: [Read original article]() Author: Sarah Constant Published: 2026-09-12T11:22:06.456390Z Content type: article Language: en Sources: [Grafana Labs blog on Grafana Labs]() Topics: [Large Language Model](), [Grafana Cloud](), [debugging](), [incident](), [telemetry](), [observability ai agents]() Tags: [ai](), [ai-agents](), [analytics](), [debugging](), [grafana](), [grafana-cloud](), [incident](), [knowledge-graph](), [llms](), [logs](), [metrics](), [monitoring](), [observability](), [telemetry](), [traces]() ### AI overview The article reports Grafana Labs experiments testing whether providing AI agents with Grafana Cloud's Knowledge Graph improves incident debugging compared with raw telemetry alone. In one incident replayed 16 times per approach, Knowledge Graph context led to the correct root cause 15 times, versus once with raw telemetry. The article also describes challenges including misleading signals, unsupported confident answers, and inconsistent investigations, arguing that well-structured context matters more than context-window size alone. ### Source excerpt On the product team here at Grafana Labs, we consider AI agents our users, too. That's why we set out to test how well agents can debug incidents across the full stack, and how much better they perform with Grafana Cloud's Knowledge Graph vs. using raw telemetry alone. Our early results are promising. In one real incident we replayed 16 times each way, an agent with Knowledge Graph context found the correct root cause 15 times, compared with just once using raw telemetry alone. Along the way, we also uncovered some of the challenges that still stand in the way of reliable AI-assisted debugging, from chasing the wrong signals to confidently making things up and producing inconsistent answers. We're still early, but our findings point to an important idea. The industry's shorthand right now is that a bigger context window will lead to better outputs. Our findings suggest it's not just about more context; it's about structuring your data well enough to serve the right context. Here's a look at what we've learned so far, as we continue to experiment out in the open and bring you along, the Grafana Labs way. Giving an agent access to telemetry is just the beginning Give a current-generation model like Opus 4.8 access to your raw telemetry during a live incident, and it genuinely starts to figure things out: querying metrics and logs, forming a hypothesis, and checking it. We have watched it work on our own incidents, and it does it affordably. But if you run software at scale, where uptime is business-critical and large teams share the responsibility, a better model alone doesn't get you all the way there for debugging. From analyzing how LLMs do root-cause analysis on our own infrastructure, and from speaking to our customers, we've uncovered three problems that get in the way: The further the cause is from the alert, the more likely the model is to get it wrong. An agent may confidently make things up when it doesn't have the evidence it needs. The same investigation c ## Presentation: From Retrieval to Reasoning: Building Production-Ready Agentic AI Systems with Knowledge Graphs DevFeed: [Presentation: From Retrieval to Reasoning: Building Production-Ready Agentic AI Systems with Knowledge Graphs]() Original publisher: [Read original article]() Author: Cassie Shum Published: 2026-09-12T11:00:00Z Content type: tutorial Language: en Sources: [InfoQ]() Topics: [Graphs](), [Architecture & Design](), [Retrieval Augmented Generation (RAG)](), [Code]() Tags: [agentic-ai](), [agentic-ai-architecture](), [agents](), [ai](), [ai-architecture](), [ai-ml-data-engineering](), [architecture](), [architecture-design](), [artificial-intelligence](), [code](), [development](), [engineering](), [infoq](), [knowledge-graph](), [knowledge-graphs-agentic-systems-patterns](), [large-language-models](), [ml-data-engineering](), [presentation](), [production](), [qcon-ai-boston-2026](), [qcon-software-development-conference](), [rag](), [reasoning](), [retrieval-augmented-generation](), [transcripts]() ### AI overview A presentation on using knowledge graphs as a foundation for production-ready agentic AI systems. It covers architectural patterns for context bundling, decision provenance, code as truth, and agent visibility, along with a graph-based engineering harness for feedback loops, token optimization, and reliability. ### Source excerpt Cassie Shum discusses why knowledge graphs serve as a critical foundation for agentic systems. Moving beyond basic RAG, she explains 4 practical architectural patterns: context bundling, decision provenance, code as truth, and agent visibility. She demonstrates an engineering harness built on a knowledge graph to streamline feedback loops, optimize token usage, and maintain system reliability. By Cassie Shum ## Second Brains in the AI era: still worth it? DevFeed: [Second Brains in the AI era: still worth it?]() Original publisher: [Read original article]() Published: 2026-09-06T00:00:00Z Content type: opinion Language: en Sources: [Mutt0-ds Notes]() Topics: [Artificial Intelligence](), [context](), [AI Agent](), [Graphs](), [Obsidian]() Tags: [ai](), [ai-agent](), [context](), [knowledge-graph](), [markdown](), [obsidian]() ### AI overview The article argues that maintaining a personal knowledge base, or "second brain," is more valuable in the AI era because it gives AI assistants structured, connected context. It suggests that linked notes resemble graph memory and can help agents navigate complex information, while large unstructured collections may increase confusion and hallucinations. ### Source excerpt I've been working on my second brain(s) for years now. Brains, plural, because I had to create new ones, for example when I changed jobs...... ## Building Trust in AI DevOps: Validating the Harness Knowledge Graph DevFeed: [Building Trust in AI DevOps: Validating the Harness Knowledge Graph]() Original publisher: [Read original article]() Author: Vikram Sahu Published: 2026-08-31T18:37:00Z Content type: article Language: en Sources: [Harness Blog]() Topics: [DevOps](), [Testing](), [Graphs](), [Artificial Intelligence](), [Software]() Tags: [ai](), [ai-evals](), [api](), [automated](), [data](), [devops](), [evals](), [graph](), [knowledge-graph](), [lifecycle](), [operational](), [other](), [pipeline](), [schema](), [sdlc](), [security](), [services](), [software](), [software-delivery](), [validation]() ### AI overview This article explains how Harness validates answers from its SDLC Knowledge Graph. Its multi-layered approach combines AI evaluations, schema traversal, API checks, direct product verification, production data, and shift-left testing to improve reliability. ### Source excerpt Discover our multi-layered validation approach combining AI evals to ensure reliable AI-powered software delivery insights. | Blog ## KV vs Prefix vs Prompt vs Semantic Caching DevFeed: [KV vs Prefix vs Prompt vs Semantic Caching]() Original publisher: [Read original article]() Author: Avi Chawla Published: 2026-08-27T20:01:47Z Content type: article Language: en Sources: [Daily Dose of Data Science]() Topics: [Caching](), [Large Language Model](), [Neo4j]() Tags: [caching](), [embedding](), [knowledge-graph](), [latency](), [llm](), [memory](), [nvme](), [object-storage](), [retrieval](), [server](), [token](), [vector]() ### AI overview The article distinguishes KV caching, prefix caching, prompt caching, and semantic caching in LLM systems. It explains what each cache stores, how exact-match and fuzzy-match behavior differ, and how cache misses affect cost, latency, and correctness. It also describes Konig's tiered storage approach for many small knowledge graphs, though the supplied text is truncated. ### Source excerpt ...explained with best practices in production. ## AI-powered clinical trial eligibility and safety using Amazon Bedrock AgentCore DevFeed: [AI-powered clinical trial eligibility and safety using Amazon Bedrock AgentCore]() Original publisher: [Read original article]() Author: Sachin Jain Published: 2026-08-19T13:11:51Z Content type: tutorial Language: en Sources: [AWS Architecture Blog]() Topics: [AI Bots]() Tags: [advanced-300](), [agent](), [agents](), [ai](), [ai-agents](), [amazon](), [amazon-bedrock](), [amazon-bedrock-agentcore](), [aws](), [data](), [how-to](), [knowledge-graph](), [llm](), [reasoning](), [safety](), [thought-leadership]() ### AI overview An AWS architecture tutorial for a clinical-trial eligibility and safety agent. It uses Amazon Bedrock AgentCore, AWS HealthLake, knowledge-graph evidence, evaluations, and human-in-the-loop review to support clinician-led screening decisions. ### Source excerpt AI agents built on Amazon Bedrock AgentCore help clinical trial teams make fast, accurate enrollment decisions while keeping clinicians in control. This post shows how to architect an eligibility and safety screening agent using AWS HealthLake, AgentCore, and AgentCore Evaluations. ## What an Ontology for AI Agents Actually Needs DevFeed: [What an Ontology for AI Agents Actually Needs]() Original publisher: [Read original article]() Author: Ananth Packkildurai Published: 2026-08-14T12:38:18Z Content type: opinion Language: en Sources: [Data Engineering Weekly]() Topics: [Artificial Intelligence](), [Graphs](), [Structured-data](), [data]() Tags: [ai](), [ai-agents](), [data](), [graph](), [knowledge-graph](), [schema]() ### AI overview The article argues that an ontology for AI agents should be treated as a governed semantic system rather than a single file or graph. It distinguishes ontology meaning from knowledge-graph facts and explains how semantic capabilities support retrieval, planning, action, verification, and operational governance. ### Source excerpt How to think about the semantic system that makes an agent coherent, governable, and useful. ## Philip Rathle on why AI agents keep reaching for a knowledge graph DevFeed: [Philip Rathle on why AI agents keep reaching for a knowledge graph]() Original publisher: [Read original article]() Author: WorkOS Published: 2026-08-05T23:21:28Z Content type: article Language: en Sources: [WorkOS Blog]() Topics: [Artificial Intelligence](), [Neo4j](), [Graphs](), [graph-database](), [data]() Tags: [agents](), [ai](), [ai-agents](), [data](), [graph](), [graph-database](), [knowledge-graph]() ### AI overview WorkOS CEO Michael Grinich interviews Neo4j CTO Philip Rathle about using knowledge graphs as an AI knowledge layer. Rathle argues that agents need deterministic, explainable access to structured company data for questions where accuracy, sovereignty, and access controls are critical, while noting that graph retrieval can provide context for better model decisions. ### Source excerpt Neo4j CTO Philip Rathle on why over 70% of new business is now the AI knowledge layer, where agents need deterministic answers, and how error rates compound. ## Crowdsourced taxonomy verification: A feedback-driven framework for refining knowledge graph relationships via online search interactions DevFeed: [Crowdsourced taxonomy verification: A feedback-driven framework for refining knowledge graph relationships via online search interactions]() Original publisher: [Read original article]() Author: Junpeng Niu Published: 2026-07-30T00:00:00Z Content type: article Language: en Sources: [Grab Tech]() Topics: [AI search](), [AI, ML & Data Engineering](), [Language models]() Tags: [algorithms](), [data](), [engineering](), [graphs](), [knowledge-graph](), [large-language-models-llms](), [llm](), [machine-learning](), [search](), [validation]() ### AI overview The article presents a feedback-driven system for verifying knowledge-graph relationships through live search interactions. It tests unverified parent-child or sibling edges in search suggestions, measures user engagement, and uses greedy exploration to balance data collection with result quality. ### Source excerpt Introduction The efficacy of semantic search relies on the accuracy of the underlying Knowledge Graph (KG). In high-velocity domains like on-demand food delivery or e-commerce, the catalog of entities like dishes, products, and merchants changes rapidly. Current methods for KG construction and maintenance face three critical challenges: Inaccuracy and hallucination from Large Language Models (LLMs): Automated models often infer relationships based on statistical text co-occurrence rather than semantic reality. For instance, an LLM might incorrectly classify "Pho" as a child of "Italian Noodle Soup" due to linguistic similarity, leading to irrelevant search results. Scalability limits of manual verification: Traditional verification relies on human annotators or domain experts. This approach is slow, expensive, and unable to keep pace with dynamic catalogs containing millions of entities. For example, daily changes in restaurant menus or grocery stock keeping units (SKUs). Error propagation in ranking: Inaccurate graph edges propagate errors downstream. If a parent-child relationship is wrong, query expansion algorithms will retrieve irrelevant items, directly degrading Click-Through Rate (CTR) and user trust. We introduce a feedback-driven verification engine that operationalizes the search interface as a validation environment. Key contributions include: User feedback-driven verification: The system treats unverified graph edges as hypotheses. Instead of accepting them as truth, it tests them against live traffic by injecting them into search suggestions and measuring user engagement. Hierarchical relationship refinement: Unlike systems that only validate entities (nodes), this framework validates structural links (edges). It confirms whether entity A is truly a parent, child, or sibling of entity B, ensuring structural integrity. Adaptive exploration: The system employs a greedy exploration policy. It intelligently balances exploitation by showing known good resul ## How Rapidflare built a million+ document ingestion pipeline for Agents on Temporal DevFeed: [How Rapidflare built a million+ document ingestion pipeline for Agents on Temporal]() Original publisher: [Read original article]() Author: Vasanth Asokan Published: 2026-07-08T00:00:00Z Content type: article Language: en Sources: [Temporal Blog]() Topics: [Agent Harness](), [Documentation](), [reliability](), [data](), [Graphs]() Tags: [agent-harness](), [agents](), [community](), [knowledge-graph](), [pipeline](), [pipelines](), [reliability](), [scale](), [technical]() ### AI overview Rapidflare explains how it uses Temporal to build durable, observable document-ingestion pipelines for technical sales agents processing knowledge bases at million-document scale. The article describes the need to ingest and pre-structure extensive technical literature into proprietary knowledge-graph formats, while preserving completeness and reliability. ### Source excerpt Rapidflare explains how it uses Temporal to run durable, observable document ingestion pipelines for technical sales Agents at million-document scale. ## From Harness Lock-In to Portable Context Layer DevFeed: [From Harness Lock-In to Portable Context Layer]() Original publisher: [Read original article]() Author: Paul Iusztin Published: 2026-06-30T08:20:17Z Content type: article Language: en Sources: [Decoding ML]() Topics: [Artificial Intelligence](), [MCP Server](), [Graphs](), [AI search](), [Notion](), [Obsidian](), [Google]() Tags: [ai](), [claude-code](), [code](), [knowledge-graph](), [mcp](), [memory](), [open-source](), [vector-search]() ### AI overview The article argues that users should keep their context layer separate from any AI harness. It proposes unified memory combining a filesystem, BM25 keyword search, semantic vector search, and a typed knowledge graph, exposed through an MCP server or skills so different models and harnesses can use it. ### Source excerpt Build a unified memory (knowledge graph or an LLM wiki) and serve it over MCP servers or skills, so any agent, open or closed, plugs in within minutes. ## Knowledge graph retrieval-augmented generation (RAG): structured retrieval for AI agents DevFeed: [Knowledge graph retrieval-augmented generation (RAG): structured retrieval for AI agents]() Original publisher: [Read original article]() Author: Jim Allen Wallace Published: 2026-06-24T00:00:00Z Content type: tutorial Language: en Sources: [Redis Blog]() Topics: [Retrieval Augmented Generation (RAG)](), [Graphs](), [AI search](), [Large Language Model]() Tags: [ai](), [embedding](), [graphs](), [knowledge-graph](), [rag](), [retrieval](), [tech-de]() ### AI overview This guide explains knowledge graph RAG as a structured retrieval approach for AI agents. It contrasts graph-based retrieval of connected entities and relationships with vector RAG's nearest-neighbor retrieval of isolated text chunks, and outlines how LLMs can build graphs from unstructured data to support multi-hop retrieval. ### Source excerpt A user asks your support agent: "is the slow-sync bug from my last ticket fixed in the version you told me to upgrade to?" Answering means connecting three documents: the customer's earlier ticket, the new release notes, and the engineering issue the ... ## How to Keep Your AI Agent's Knowledge Graph Clean DevFeed: [How to Keep Your AI Agent's Knowledge Graph Clean]() Original publisher: [Read original article]() Author: Paul Iusztin Published: 2026-06-02T05:00:53Z Content type: tutorial Language: en Sources: [Decoding ML]() Topics: [Entity resolution](), [Graphs](), [AI Agent](), [data](), [Large Language Model](), [Neo4j]() Tags: [agent](), [ai-agent](), [code](), [knowledge-graph](), [llm](), [pipeline](), [rag](), [review]() ### AI overview This tutorial explains how to keep an AI agent's knowledge graph clean by separating entity resolution for naming from deduplication for identity. It presents a pipeline covering LLM extraction, embeddings, deduplication, and safety reviews to reduce incorrect merges as the graph grows. ### Source excerpt The resolution, deduplication, and review pipeline that keeps agent memory usable as it grows. ## How to Start a Knowledge-Graph Memory Layer with an Extendable Ontology DevFeed: [How to Start a Knowledge-Graph Memory Layer with an Extendable Ontology]() Original publisher: [Read original article]() Author: Paul Iusztin Published: 2026-05-26T05:00:51Z Content type: tutorial Language: en Sources: [Decoding ML]() Topics: [Graphs](), [Retrieval Augmented Generation (RAG)](), [data]() Tags: [ai-engineering](), [article](), [how-to](), [knowledge-graph](), [knowledge-graphs](), [rag](), [schema-design]() ### AI overview This tutorial explains how to begin building a knowledge-graph memory layer without designing a complete ontology upfront. It recommends a small, generic, extendable POLE+O model with Preferences and Facts, then extending it as domain-specific data reveals gaps. ### Source excerpt Start with a fixed, generic base and extend only when your data demands it. ## Scaling Airbnb's identity graph with a unified knowledge graph infrastructure DevFeed: [Scaling Airbnb's identity graph with a unified knowledge graph infrastructure]() Original publisher: [Read original article]() Author: Lucen Zhao Published: 2026-05-19T17:01:01Z Content type: article Language: en Sources: [The Airbnb Tech Blog - Medium]() Topics: [Graphs](), [graph-database](), [Architecture & Design](), [Scalability](), [Caching](), [Low Latency](), [real-time](), [data]() Tags: [architecture](), [caching](), [data](), [engineering](), [graph-database](), [infrastructure](), [knowledge-graph](), [latency](), [low-latency](), [safety](), [scalability](), [technology]() ### AI overview Airbnb describes scaling its identity graph into a unified, internally managed knowledge graph infrastructure. The article covers the graph database and KV caching storage layer, graph service, near-real-time ingestion, low-latency access, and architectural evolution driven by increasing graph density and query complexity. ### Source excerpt How Airbnb shifts from PaaS to an internal knowledge graph infrastructure at scale. By: Lucen Zhao, Shukun Yang, Ashish Jain Knowledge graphs offer a natural and powerful way to represent relationships between entities. Many real-world systems are fundamentally about connections. Airbnb's identity graph captures relationships between users in a graph database. The identity graph serves aggregated insights that enable user identity resolution and relationship understanding. These capabilities support a wide range of Trust and Safety use cases, from detecting suspicious activities to identifying linked accounts. Over time, the identity graph has grown into one of the largest and most complex graph data products at Airbnb, both in terms of scale and the complexity of queries it supports. In 2024, Airbnb began investing in a new, internally managed, paved-path graph data platform to build a unified knowledge graph infrastructure. Airbnb's identity graph became one of the first systems to adopt this platform. In this post, we'll walk through the foundations and challenges of the identity graph, introduce the architecture behind the graph infrastructure, and highlight several key optimizations that emerged during the onboarding process. Airbnb's identity graph Airbnb's identity graph is a critical foundation layer, playing an important role in Trust and Safety applications. It contains two major components: Graph data storage: a storage layer composed of a graph database and a key-value (KV) caching layer. It models users and relationships as vertices and edges. Most data is ingested in near real-time through asynchronous events and served through low-latency, real-time service calls. Graph service: this service provides a unified interface for accessing graph data. It retrieves data from underlying sources, including the graph database, applies aggregation logic or models as needed, and serves the results to downstream customer services. Evolution of the identity graph a ## Inside Neo4j's Agent Memory DevFeed: [Inside Neo4j's Agent Memory]() Original publisher: [Read original article]() Author: Paul Iusztin Published: 2026-05-19T08:55:51Z Content type: tutorial Language: en Sources: [Decoding ML]() Topics: [Neo4j](), [Graphs](), [Artificial Intelligence](), [Retrieval Augmented Generation (RAG)]() Tags: [agent](), [ai](), [graph](), [knowledge-graph](), [memory](), [rag](), [retrieval-augmented-generation-rag]() ### AI overview The article presents Neo4j knowledge graphs as a model for durable agent memory. It argues that file-based logs and vector indexes lack identity and relationship tracking, while a structured graph can connect entities, preferences, and facts across a growing knowledge base. ### Source excerpt The knowledge-graph patterns that turn one-shot conversations into compounding intelligence. ## Proofs and Essays Are Paths: An LLM ↔ Prover Loop for Falsifying Hallucinations DevFeed: [Proofs and Essays Are Paths: An LLM ↔ Prover Loop for Falsifying Hallucinations]() Original publisher: [Read original article]() Published: 2026-04-30T00:00:00Z Content type: opinion Language: en Sources: [Alex Korbonits]() Topics: [Large Language Model](), [Automated reasoning](), [Graphs](), [Artificial Intelligence](), [Math and Logic]() Tags: [grounding](), [hallucinations](), [knowledge-graph](), [llm](), [reasoning](), [smt](), [verification](), [verify]() ### AI overview This essay proposes a bidirectional loop between a large language model and a theorem prover. LLM outputs become propositions and graph connections that the prover checks, while verification results feed back into the model. In this framework, hallucinations appear as knowledge-graph paths with edges that fail formal verification. ### Source excerpt If a proof is a path through a formal space and an essay is a path through a semantic one, can we close the loop -- translate LLM outputs into a theorem prover, build a knowledge graph of verifiable propositions, and use the gaps as a signal for hallucination? Notebook ideas from March 2024 expanding on the SMT-grounding piece. ## I Built the Knowledge Graph of Machine Learning DevFeed: [I Built the Knowledge Graph of Machine Learning]() Original publisher: [Read original article]() Author: Tivadar Danka Published: 2026-04-19T07:46:57Z Content type: article Language: en Sources: [The Palindrome]() Topics: [Machine Learning & Artificial Intelligence](), [structure]() Tags: [knowledge-graph](), [machine-learning]() ### AI overview The article explores the structure of machine learning and presents a knowledge graph of the field. ### Source excerpt Exploring the structure of machine learning ## Introducing the Vercel plugin for coding agents DevFeed: [Introducing the Vercel plugin for coding agents]() Original publisher: [Read original article]() Author: Andrew Qu Published: 2026-03-17T13:00:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [AI Bots]() Tags: [agents](), [bash](), [claude-code](), [codex](), [coding](), [cursor](), [deployment](), [knowledge-graph](), [next-js](), [retrieval](), [skills](), [tooling](), [validation](), [vercel]() ### AI overview Vercel announces a plugin for Claude Code and Cursor that injects project and platform knowledge into coding-agent context. It provides skills, specialist agents, slash commands, context management, and real-time validation. ### Source excerpt Claude Code and Cursor can now further understand Vercel projects using the new Vercel plugin and a full platform knowledge graph. The plugin observes real-time activity, including file edits and terminal commands, to dynamically inject Vercel knowledge into the agent's context. Key capabilities include: Platform knowledge: Access 47+ skills covering the Vercel platform, including Next.js, AI SDK, Turborepo, Vercel Functions, and Routing Middleware, powered by a relational knowledge graph Specialized tooling: Use three specialist agents (AI Architect, Deployment Expert, Performance Optimizer) and five slash commands (/bootstrap, /deploy, /env, /status, /marketplace) Context management: An injection engine and project profiler rank, deduplicate, and budget-control loaded context Code validation: PostToolUse validation catches deprecated patterns, sunset packages, and stale APIs in real time Instead of standard retrieval, the plugin compiles pattern matchers at build time and runs a priority-ranked injection pipeline across seven lifecycle hooks. Skills fire when glob patterns, bash regexes, import statements, or prompt signals match, and are then deduplicated across the session to ensure accurate agent responses. The plugin currently supports Claude Code and Cursor, with OpenAI Codex support coming soon. Install the plugin via npx: Directly in Claude Code via the official marketplace: Or directly in Cursor: Explore the source code in the Vercel plugin repository. Read more ## Reinventing Risk at Revolut DevFeed: [Reinventing Risk at Revolut]() Original publisher: [Read original article]() Author: Konstantin Vasilev Published: 2024-12-10T07:46:23Z Content type: opinion Language: en Sources: [Revolut Engineering]() Topics: [Finance](), [Artificial Intelligence](), [Software]() Tags: [ai](), [banking](), [fake-accounts](), [finance](), [financial](), [fraud](), [knowledge-graph](), [operations](), [outage](), [revolut](), [risk-management](), [services]() ### AI overview Revolut describes how its Risk function approaches financial and operational risk in a financial-services business. The article explains that risks can overlap, such as fraud involving fake accounts leading to unpaid loans, and discusses the limitations of segregating risks into separate departments. ### Source excerpt Using probability, reasoning and AI Revolut is known for its innovation in delivering "super-app" financial services to its customers. The success is evident not only through the rocketing user growth, but also via the ripple effects caused across the banking industry -- traditional finance has to reinvent and evolve, or remain part of history, just like the internet search replaced the paper phonebook. Behind this success stands a complex and impressive machinery, which is still piloted by the same founders, who also engineered the very first components and set the Revolut rocketship in motion. This machine is mostly known for its software excellence, however, with any financial services comes the responsibility of navigating a complex net of regulations and risks. True to the company values, the Risk function at Revolut also innovates, leveraging the decades of banking experience of its leadership and adding the challenger way of thinking with the latest technology. What is Risk? One can find many definitions of risk, ranging from explanations suitable for children, "The possibility of something bad happening", to advanced statistical definitions used in finance: Uncertainty of the return on investment, characterised by standard deviation, asymmetry and tails of the return distribution and the generic ISO 31000 definition: "The effect of uncertainty on objectives" In addition, both graduate textbooks and banking regulations define two general types of risk: Financial: related to financial positions, such as: user deposits, lending, and investments in market instruments. Operational: related to failure of operations, such as: system outage, fraud and natural disasters. Some of you might already sense the inevitable challenges in distinguishing the above definitions and risks. Events seldom happen in isolation and independently: a natural disaster can lead to system outages and market crashes, similarly fraudsters might use fake accounts to withdraw loans which would ## From Pre-trained to Fine-tuned: Nextdoor's Path to Effective Embedding Applications DevFeed: [From Pre-trained to Fine-tuned: Nextdoor's Path to Effective Embedding Applications]() Original publisher: [Read original article]() Author: Karthik Jayasurya Published: 2023-09-07T11:31:32Z Content type: article Language: en Sources: [Nextdoor]() Topics: [Embeddings](), [Fine-tuning](), [Machine learning](), [Natural language processing](), [Transformer](), [Development](), [recommendation systems]() Tags: [deep-learning](), [development](), [embedding](), [embeddings](), [knowledge-graph](), [machine-learning](), [ml](), [models](), [ranking](), [real-time](), [recommendation-system](), [systems]() ### AI overview Nextdoor describes its evolution from using pre-trained transformer models as embedding feature extractors to fine-tuning embeddings with unlabelled and labeled data. The article covers how embeddings are developed, featurized, and served at scale for applications including notification scoring and feed ranking. ### Source excerpt Background The majority of ML models at Nextdoor are typically driven by a large number of features that are primarily either continuous or discrete in nature. The personalized features usually stem from historical aggregations or real-time summarization of interaction features, typically captured through logged tracking events. However, representing content through deep understanding using information behind it (text/image) is crucial for modeling nuanced user signals and better personalizing complex user behavior across many of our products. In the rapidly evolving field of NLP, utilizing transformer models to perform representation learning effectively and efficiently has become increasingly important for user understanding and improving their product experience. Towards that, we have built a lot of entity embedding models spanning entities such as posts, comments, users, search queries & classifieds. We first leveraged deep understanding of content and used that to derive embeddings for meta entities like users based on their past interacted content. These powerful representations are found to be very crucial towards extracting meaningful features for some of the biggest ML ranking systems at Nextdoor such as notifications scoring and feed ranking. By making them readily available and building to scale, we can drive adoption of state-of-the-art reliably and put them in the hands of ML Engineers for rapidly building performant models across the company. This blog primarily focuses on how we iterated on the development of embedding models, how they are featurized and served at large scale into various product applications as well as some of the challenges encountered during this process. We summarize the evolution of work across three sections. In section 1, the focus is to leverage state-of-the-art pre-trained models to rapidly evaluate the value of embeddings models as feature extractors. Section 2 describes how to fine-tune embeddings using unlabelled data for ## Stanford AI Lab Papers and Talks at ICLR 2022 DevFeed: [Stanford AI Lab Papers and Talks at ICLR 2022]() Original publisher: [Read original article]() Author: Compiled by Drew A. Hudson Published: 2022-04-25T07:00:00Z Content type: article Language: en Sources: [The Stanford AI Lab Blog]() Topics: [Artificial Intelligence](), [Benchmark](), [benchmarking](), [Reinforcement learning](), [dataset](), [Language models](), [Graphs](), [Robotics](), [Robotic manipulation]() Tags: [benchmark](), [benchmarking](), [datasets](), [iclr](), [knowledge-graph](), [language-models](), [neural-networks](), [reinforcement-learning](), [robotic-manipulation](), [robotics]() ### AI overview Stanford AI Lab presents its work at ICLR 2022, including papers and talks on reinforcement learning, benchmark datasets, distribution shifts, in-context learning, language models, graph reasoning, model editing, robotics, and robotic manipulation. ### Source excerpt The International Conference on Learning Representations (ICLR) 2022 is being hosted virtually from April 25th - April 29th. We're excited to share all the work from SAIL that's being presented, and you'll find links to papers, videos and blogs below. Feel free to reach out to the contact authors directly to learn more about the work that's happening at Stanford! List of Accepted Papers Autonomous Reinforcement Learning: Formalism and Benchmarking Authors: Archit Sharma*, Kelvin Xu*, Nikhil Sardana, Abhishek Gupta, Karol Hausman, Sergey Levine, Chelsea Finn Contact: architsh@stanford.edu Links: Paper | Website Keywords: reinforcement learning, continual learning, reset-free reinforcement learning MetaShift: A Dataset of Datasets for Evaluating Contextual Distribution Shifts and Training Conflicts Authors: Weixin Liang, James Zou Contact: wxliang@stanford.edu Links: Paper | Video | Website Keywords: benchmark dataset, distribution shift, out-of-domain generalization An Explanation of In-context Learning as Implicit Bayesian Inference Authors: Sang Michael Xie, Aditi Raghunathan, Percy Liang, Tengyu Ma Contact: xie@cs.stanford.edu Links: Paper | Video Keywords: gpt-3, in-context learning, pretraining, few-shot learning GreaseLM: Graph REASoning Enhanced Language Models for Question Answering Authors: Xikun Zhang, Antoine Bosselut, Michihiro Yasunaga, Hongyu Ren, Percy Liang, Christopher D. Manning, Jure Leskovec Contact: xikunz2@cs.stanford.edu Award nominations: Spotlight Links: Paper | Website Keywords: knowledge graph, question answering, language model, commonsense reasoning, graph neural networks, biomedical qa Fast Model Editing at Scale Authors: Eric Mitchell, Charles Lin, Antoine Bosselut, Chelsea Finn, Christopher D. Manning Contact: eric.mitchell@cs.stanford.edu Links: Paper | Website Keywords: model editing; meta-learning; language models; continual learning; temporal generalization Vision-Based Manipulators Need to Also See from Their Hands Authors: Kyle H [Next page]()