iCreatAI News
Back to main site
Skip to content
#Agent
Nov 3Tue
  1. Claude Code Releases20

    v2.1.317

    Claude Code v2.1.317 adds a warning to /schedule when a new cloud scheduled task receives connectors you did not request, a confirmation before a web or shell tool call sends your account email address to the network, and onFailure: "block" for Agent SDK hook callbacks. It also adds anthropic.plan, anthropic.seat_type, anthropic.provider and anthropic.custom_base_url to OTel metrics, events and spans, plus a cost report endpoint at GET /v1/organizations/analytics/cost on the Claude apps gateway admin API. The release includes numerous fixes across MCP servers, auto mode, self-hosted runners, OpenTelemetry exports, VSCode, cloud sessions and Remote Control.

  2. Latent Space40

    Building Legal Agents When You Can't See the Data - Jacob Lauritzen, CTO of Legora

    In this podcast episode, Legora CTO Jacob Lauritzen discusses how the legal-tech company builds and tests agents whose traces it cannot read because the data belongs to law firm clients, using simulated test environments modeled on customer work. He covers model choices and the addition of a model picker, an estimate that models already do about 60% of legal work well, and a citation fix that took accuracy from 70% to 99% in one release. The conversation also touches on Tabular Review, running everything on Azure, forward-deployed and legal engineers, and the future of law firm partnerships, the billable hour, and junior lawyer roles.

  3. Latent Space60

    Building AI for Lawyers - Jacob Lauritzen, CTO of Legora

    Jacob Lauritzen, CTO of Legora, discusses building AI agents for lawyers after the company's $550M Series E at an $8.8B valuation, covering how its evals team simulates customer work with three engineers and three lawyers because client data is confidential, and how Legora runs everything on Azure for law-firm approval. He says models already handle about 60% of legal work well, that complex analysis and editing across million-token contracts still need more intelligence, and that Legora has begun post-training open-weight models for narrow tasks like legal research. The company passed $100 million in ARR in April, less than 18 months after launch, and now has more than 1,800 customers.

  4. TechCrunch · AI55

    Trajectory raises $45M to make AI model training accessible to nontechnical experts

    Trajectory, an 18-person startup founded by three Stanford classmates with experience at Google DeepMind and Apple, raised $45 million in a Series A round at a $300 million valuation led by Sequoia, with participation from Nvidia, Bessemer Venture Partners, and Bain Capital Ventures. The company uses continual learning to let businesses connect an open-source model to their applications, collect feedback on its performance, and retrain it so specialists like lawyers and doctors can improve models themselves. Customers including Harvey, Rogo, Decagon, Mercor, and Nvidia are already using the platform, which is now publicly available.

  5. Hacker News AI43

    CloudOpsBench: AI benchmark shows FAI models struggling with cloud infra

    CloudOpsBench is a benchmark for AI agents on real cloud operations, and its leaderboard ranks models by resolution rate, harness, release date, time, tokens and cost. The top entry, claude-fable-5-1 (max) with claude-code, reaches a 60.6% resolution rate at 38.6h, 97.9M tokens and $404.3, while the lowest listed, kimi-k3.1-pro (high) with terminus-2, reaches 19.1%.

  6. Hacker News AI22

    AI and Incidents with Sylvain Kalache [video]

    In this Arrested DevOps podcast episode, host Matty Stratton talks with Sylvain Kalache, who leads an AI lab at incident management platform Rootly, about how AI-assisted incident response tools (often called "AI SRE") combine long-term memory such as codebases, incident reports and Slack history with live signals to generate and investigate hypotheses. They discuss where these tools excel on simple incidents versus complex ones, the need to keep humans in the loop, and the risk that automating away easy incidents erodes engineers' troubleshooting skills, drawing a parallel to aviation's autopilot and simulator training.

  7. Hacker News AI34

    AI × Hiring

    The authors interviewed 30 recruiting and engineering leaders in New York and San Francisco in August and September 2026 about how AI is changing hiring, covering inbound, sourcing, referrals, evaluation and winning candidates. They report that fraud can make up 25% to 80% of applications for remote roles, that no team trusts AI to rank applicants, and that agentic sourcing output degrades after the first two to three days unless someone keeps steering it. The report also describes new evaluation formats such as system design, building with AI and reviewing AI-written code, and argues that effective filters should be harder for low-quality candidates to fake than for strong ones to emit.

  8. Hacker News AI58

    Lyngvi – A Hardened Sandbox for AI

    Lyngvi is an MIT-licensed, agent-agnostic sandbox for running autonomous coding agents against your own repositories on macOS and Linux via Podman, driven from a JavaFX desktop app or a CLI that share one Java core. It gives each conversation its own container with no default network route (egress only through an allowlisted Squid proxy), a copied working tree with a fresh .git, and secrets held in a gateway sidecar, with changes returned as host-classified patches (apply, hold, reject) committed only to a working branch. It ships reference adapters for Claude Code and OpenCode, supports local model servers through the proxy, and pins its base images by digest.

  9. Sierra Blog58

    Redesigning Agent Studio around Ghostwriter

    Sierra has redesigned Agent Studio around Ghostwriter, its always-on agent for building agents, so users can investigate issues, edit journeys, run simulations, and test fixes within a single conversation. Ghostwriter now follows users through Agent Studio and sees what they are viewing, combining capabilities previously spread across Explorer, Sierra Assistant, and other specialized agents, with each piece of work kept in its own shareable or forkable session. The redesign also brings related agents into a single project, adds line-by-line change comparisons, Preview Chat and Simulations, and release governance with required approvals and controlled rollouts.

  10. Pydantic AI Articles38

    The best Python AI agent frameworks in 2026

    This roundup compares nine Python agent frameworks against criteria including model portability, durable execution, human approval, frontend and voice reach, MCP support, tracing and commercial support, and picks Pydantic AI overall for covering all of them in the open-source library. It then recommends LangGraph/LangChain, the OpenAI Agents SDK, Google ADK, Microsoft Agent Framework, Strands Agents, CrewAI, the Claude Agent SDK and Agno for specific vendor or design preferences, each with a stated catch. A comparison table lists model providers, durable execution engines, speech-to-speech voice, interfaces, MCP, observability, commercial support and license per framework, checked October 29, 2026.

  11. Hacker News AI34

    Hercules Founder Brendan Falk on the AI Opportunity

    Brendan Falk, who co-founded the developer tools startup Fig and sold it to AWS in 2023, now runs Hercules, a seven-person company he says is doing tens of millions in revenue by letting non-technical people build production-grade business software such as ERPs and CRMs. He describes Hercules's bottoms-up, milestone-based approach, its use of low-CAC Google Ads in emerging markets for product feedback, and his view that custom software is a revenue-maximizing strategy rather than a cost-saving one.

  12. Hacker News AI60

    AI evals for agents: what they are and how to write them

    This guide defines the vocabulary of AI evals (task, trial, grader, transcript, outcome, suite) and explains how agent evals differ from LLM evals: they grade a whole run, including tool calls, arguments, resulting state and final reply, not just one response. It covers the three grader types (code, model/LLM judge, human), the distinction between capability and regression evals, and pass@k versus pass^k, then walks through writing an agent eval in seven steps with a worked refund example.

  13. Hacker News AI38

    Secure AI Agents with Hazmat Suits, Not Self-Defense Classes

    The article argues that AI agents, being powered by probabilistic LLMs, cannot reliably recognize risky instructions, so behavioral safeguards like system-prompt rules are insufficient and architectural safeguards are needed. It describes emerging infrastructure—sandboxes, credential brokers that attach secrets on egress via an HTTPS_PROXY, and gateways for granular permissions—that deterministically prevent harmful actions. The piece is published by Infisical, which promotes its Agent Vault product at the end.

  14. The Verge · AI40

    Anthropic clamps down on Claude’s AI video game clones

    Anthropic updated Claude Code's system prompt to stop the tool from recreating existing video games, while still allowing users to build original games, games in a genre, and games from their own designs and assets. Anthropic's Addy Osmani said the company added two sentences to the prompt on Opus 5.5 because using Claude to infringe IP violates its usage policy, and that it will keep iterating to strike the right balance. Anthropic was asked whether it has or will update other Claude system prompts as well.

  15. Hacker News AI38

    AI assistants read a Shopify store 10x more often than people click through

    A 30-day analysis of a headless Shopify fashion storefront found that ChatGPT and Claude fetched pages 1,264 times for users while GA4 recorded only 122 visits from AI answers, with ChatGPT alone reading pages about nine times per visit. The logs also showed assistants repeatedly requesting discontinued products that returned soft 404s, Claude probing standard Shopify paths the storefront doesn't serve, and roughly 10,600 requests from fake AI crawlers that failed IP verification and mostly probed for secrets. The authors note their own dashboard initially overstated ChatGPT arrivals by more than tenfold due to utm tag handling in referer headers, and recommend checking 404 status codes, adding redirects for standard Shopify paths, and verifying bots by IP.

  16. Hacker News AI18

    Show HN: CpuTopBars and devTopBars – macOS menu bar system and AI monitors

    devTopBars is a paid macOS menu-bar app built on the free, open-source cpuTopBars, adding bars for Claude Code's 5-hour and weekly usage limits and for each live session, plus LAN/WAN/VPN address rows and a keep-awake option. It requires macOS 27 or later and a free open-source usage-feed plugin installed once in Claude Code, and it is currently in open beta via TestFlight ahead of an App Store release. The app sends no session or user data anywhere, making only an ipify WAN lookup on click and a per-launch GitHub check for plugin updates.

  19. Claude Blog60

    Make long-running agents faster with message threads

    Anthropic announced that message threads is in public beta on the Claude Platform, letting agents send only new messages plus the previous response ID instead of resending full conversation history each turn. With message threads on by default for most Claude Code users, requests of 500k to 1M tokens saw median time to first token drop by about 35%, though token usage stays the same since Claude still reads the full conversation each turn. Threads are held encrypted in memory, expire 5 minutes after the last request, and are available on the native Claude Platform, Claude Platform on AWS, Claude in Microsoft Foundry, and Claude Models on Google Cloud (Anthropic managed).

    Why it matters: It gives developers concrete latency figures and setup details for a new API feature aimed at long-running agent sessions.

  20. Hacker News AI22

    From Cloud-First to AI-First: Productivity-Maxxing or Token-Maxxing?

    The article contrasts a 2016 cloud migration debate, where an engineer had to justify the move to skeptical executives, with a 2026 AI push where budget and permission arrive before the experiment is designed. It argues that engineers' questions about review capacity, permissions, and data access are useful inputs to a pilot, and that teams should start with a neglected migration, track review and rework costs alongside token spend, and keep tests, conventions, and customer observations from each run. Success should be judged by whether the next engineer can explain the change and the on-call person can recover from failure, not by activated seats or token usage.

  21. ElevenLabs Blog30

    How KPN is bringing voice AI to the Netherlands' entrepreneurs with Mobiel AI

    KPN partnered with ElevenLabs to build Mobiel AI, an add-on to KPN EEN MKB mobile subscriptions that offers voicemail transcription, call summaries and transcripts with follow-up actions, and call routing and reachability settings. KPN said ElevenLabs stood out for speech quality and conversational AI in Dutch, and the teams raised verification accuracy from roughly 50% to 90%. Mobiel AI runs on KPN's network and data centers, with voice data staying in the Netherlands and customer data not used to train AI models or shared with third parties.

  22. Google Blog AI22

    The latest AI news we announced in October 2026

    Google recapped its October 2026 AI updates, led by the global rollout of the Gemini 4 Argon frontier reasoning model and the announcement of Gemini 4 Carbon, its strongest model yet for long-horizon engineering and cybersecurity defense, now in pre-release testing with the U.S. government and trusted cyber defenders via the Fairwind Program. The month also brought Genie 4 for real-time interactive world simulation with native audio, the Google Fitbit Edge fitness tracker, a global expansion of SynthID Detector, and a quantum computing breakthrough in which "magic state cultivation" on the Willow processor cut errors roughly 40-fold. Other items included AMIE results published in The Lancet, the Playground gaming platform, EmbeddingGemma 2, the first AI response from a Project Suncatcher satellite, and Research@ 2026 announcements.

  23. Hacker News AI38

    A proposal: how to hire junior engineers in the age of AI

    The author proposes replacing the coding interview with a structured exercise in which candidates spend an hour or two extracting requirements from an LLM role-playing as a customer and directing a second LLM to build working software, with an agent generating tests and a human making the final judgment. He argues junior hiring has collapsed since 2023, that internships and elite-school or "hunger" filters are inadequate or biased, and that the exercise mirrors medicine's OSCE standardized-patient exam. The approach is untested, but he contends the company that solves junior hiring will gain a supply of affordable product engineers.

  24. Hacker News AI58

    Show HN: A persistent Chrome cloud browser for AI agents

    Cloudbrowser offers AI agents an isolated Chrome cloud browser with a persistent profile, a CDP URL for automation, and a live view accessible from anywhere. Users can pause, resume, and fork browser profiles, run concurrent sessions, and connect tools like Claude Code, Codex, Cursor, and Playwright. The Free plan includes 1 browser-hour per month and one running browser, while Basic ($20/month) offers 250 hours and 10 concurrent browsers and Pro ($99/month) offers 1,250 hours and 100 concurrent browsers.

  26. Hacker News AI34

    Sunscreen, a stylus, and 16 hours of pair programming with an AI

    A kernel developer describes using Claude to add handwriting recognition to KDE's plasma-keyboard on-screen keyboard, after finding Qt Virtual Keyboard's handwriting mode is gated behind a compile-time flag requiring proprietary engines. Over roughly 15–16 hours of active work he produced four commits (~5,000 lines across 32 files) including a handwriting mode, editing keys, scribble-to-correct, and 79 tests, using PhatWare's GPL-3 WritePad engine (6.75% character error rate) after a self-trained model failed at 82.93%. He credits the AI with speed and fluency across unfamiliar stacks, but argues a human was load-bearing for catching wrong assumptions, insisting on measurement over plausible fixes, and product judgment.

  27. Hacker News AI58

    Show HN: ArchDocs – Shared context and verification for engineers and AI agents

    ArchDocs is a tool that connects design docs, architecture decisions, implementation plans and recorded work so that engineers and coding agents work from the same plan, with agents accessing specs and open questions over MCP. It compares planned changes against recorded changes and GitHub pull requests and checks to surface gaps in both directions, and keeps a work log with handoffs for resuming tasks. The Free plan includes architecture docs, MCP access and one design doc, with unlimited design docs on Team and reviews, approvals and GitHub evidence on Business and Enterprise.

  28. AWS Machine Learning Blog38

    How Circle K accelerated DevOps incident triage to under 15 minutes with Amazon Bedrock

    Circle K built an agentic AI triage system on Amazon Bedrock that investigates Grafana alerts across AWS resources, OpenSearch logs, and GitHub code, then posts a structured root-cause summary to Slack. Initially built on Amazon Bedrock Flows, the system was migrated to a single agent on Amazon Bedrock AgentCore Runtime with AgentCore Gateway brokering data source access. Circle K reports mean triage time dropped from 45–60 minutes to under 15 minutes, a 67–75 percent reduction based on internal incident-response metrics.

  29. AWS Machine Learning Blog40

    Agentic engineering intelligence with Amazon Nova Multimodal Embeddings and Amazon Bedrock AgentCore

    This post describes a multi-agent engineering assistant built on Amazon Bedrock AgentCore, Amazon Nova Multimodal Embeddings, and the open source Strands Agents SDK that assembles evidence for aerospace part substitution decisions. A planner routes queries to Catalog, Evaluation, and Knowledge specialists, with deterministic Python tolerance-rule scoring, Nova Multimodal Embeddings for multimodal retrieval and photo cosine similarity, and a supervisor that synthesizes results only when multiple specialists contribute. The engineer remains the decision-maker, and the system drafts justification memos and Integration Impact Assessments with audit trail entries aligned with AS9100 sections 8.4 and 8.5.6.

  30. AWS Machine Learning Blog36

    Build efficient agents with knowledge distillation on Amazon SageMaker AI

    This post explains how to use sequence-level knowledge distillation (SeqKD) to fine-tune a smaller shopping agent on Amazon SageMaker AI, using a frontier teacher (GLM-5.3 on Amazon Bedrock) to generate multi-step tool-use trajectories that are filtered by ShoppingBench's programmatic scorer and used for serverless supervised fine-tuning of a Qwen3.5-4B student. The distilled student reached an average Absolute Success Rate of 0.541 on 324 held-out ShoppingBench problems, versus 0.297 for the base model and 0.540 for the greedy teacher, while IFEval general instruction-following dropped roughly 6 to 8 points. The post also covers per-step trajectory encoding, LoRA training configuration, a low-cost model-evaluation pre-check, and serving the base model and student adapter as two inference components on one real-time endpoint.

  31. MIT Technology Review · AI22

    Raising the quality of insights in agentic analytics

    An MIT Technology Review Insights report finds that on average only 48% of insights generated by analytics agents at surveyed companies lead to actions that improve the business, up from 37% a year earlier, with a small group of "insight leaders" exceeding 70%. Respondents cite lack of business context (55%) and fragmented or inaccessible data (45%) as the main causes of substandard insights, and 52% of organizations say agents generate too many insights to evaluate properly. Companies are prioritizing stronger AI and data governance (60%) and adding a semantic layer to their data architecture (45%).

  32. Hacker News AI42

    Vibe-Coding a Game in 2026: What AI Builds, What Breaks

    This editorial guide argues that in 2026 AI can produce a playable small browser game in an evening, citing Simon Willison's October 28 claim that building games that play well is "shockingly easy now" and Roblox's report of over 70,000 games published with its prompt-to-game tool. It details where AI still fails — long-session game logic, consistent art and animation, multiplayer and replays, performance on mid-range Android phones, and server-side score and payment checks — and summarizes Steam, Google Play, App Store and Telegram rules for AI-made games. It closes with guidance on when to DIY versus hire a studio, noting the studio's own starting prices from $8,000 to $25,000.

  34. Hacker News AI48

    Show HN: aido - Spec-driven, AI Augmented Engineer workflow with Claude Code

    aido is a Claude Code plugin that hands the agent a whole feature—spec, tests, code, commits and pull request—while stopping at each decision that matters for the engineer to approve. It centers every stage on a SPEC.md with numbered acceptance criteria, checks the spec against the repository's CLAUDE.md and AGENTS.md, and offers a fresh-context /aido:verify to confirm the implementation matches what was agreed. Skills cover single-phase and multi-phase features plus utilities for stabilizing existing PRs, resolving migration conflicts and mapping architecture.

  35. Hacker News AI58

    We replaced 11,000 lines of AI chat code with CopilotKit and AG-UI

    fynk describes replacing its custom AI chat with the open AG-UI protocol and CopilotKit, removing about 11,000 lines of its own code. Because CopilotKit was React-only and fynk is a Vue app, engineer Abel Enekes contributed a Vue port (62,517 lines across 244 files) that was merged on 13 May 2026 and published as @copilotkit/vue. The post details the assistant's shared state, tool-call traces, human-in-the-loop confirmations, generative UI, and in-house citations and permissions, and advises adopting what every agent needs while building what differentiates the product.

  37. Hacker News AI20

    Show HN: Agent Drift Game – short games about why AI agents drift

    Agent Drift Game is a browser-based set of short games, a few minutes each with nothing to install, where after finishing you can see how AI agents played the same game and copy a prompt to try it on your own agent. Episode 1 (Direction drift) and Episode 2 (Position drift) each ask where you first say NO, and a more complex version for your own machine, involving a real AI workflow stuck at a gate with Claude Code or Codex starting from a handoff report containing one wrong judgment, is to be continued. The creators are inviting a few players to try it privately before it goes public and share what they said.

  39. Tomasz Tunguz62

    The Four Questions of AI

    The piece lays out four coupled questions across the AI stack: how application companies build moats when models can absorb their features, who owns the harness and context layer given evidence that third-party harnesses outperform native ones, when foundation-model training costs stop paying back, and whether credit and power can keep pace with GPU demand. It cites figures including 98% input-token share, a 71% cost cut from smarter harnesses, roughly 83% of agent tasks routing to smaller models, $4t in projected five-year AI data center debt, and a B300 index at $12.29 per GPU-hour. It concludes that the layers behave like a double pendulum—simple individually but chaotic when coupled.

  40. Hacker News AI42

    You Can't Just Add AI

    The article argues that simply bolting AI tools onto human-first workflows yields limited productivity gains, drawing a parallel to the electrical revolution's Productivity Paradox where factories only benefited after being redesigned around electric motors. It contends that becoming AI-native requires rebuilding systems around three properties—context (e.g., monorepos, unified observability), feedback (e.g., static typing, fast verification), and agency (CLIs, MCP servers, isolated remote compute for autonomous agents). The author shares examples from their team, including an agent that traced a Kafka client issue and another that found an AWS SDK memory regression.