Skip to content
AI NEWS JOURNAL
AI, IN PERSPECTIVE

Search AI news

Search headlines, summaries and articles to connect the dots.

Newest first

n8n@2.47.7

The n8n 2.47.7 update resolves a bug where executions using global credentials were incorrectly redacted, ensuring proper visibility of credential data in execution logs.

n8n@2.48.3

The n8n 2.48.3 update addresses several bugs related to Redis caching, prompt caching, legacy data handling, and UI visibility, introduces a new feature to display Gateway credits promotions in the editor, and includes performance enhancements by optimizing character map inclusion and reducing per-item setup in the expression runtime.

n8n@2.48.4

The n8n 2.48.4 update resolves a bug where executions using global credentials were incorrectly redacted, ensuring proper visibility of credential data in execution logs.

Optimizing cost per task and proving agent ROI on Amazon Bedrock

Comparing per-token prices hides the real cost of running a model on Amazon Bedrock. This post shows how to calculate the true cost per task across four hidden cost drivers, reduce it with caching, routing, and output constraints, and prove agent ROI with human baselines, holdout experiments, and a five-dimension scorecard.

Parallel Decisions now available on AI Gateway for free

Parallel Decisions Fast is now available on AI Gateway at no cost until November 23, 2026. It searches the live web to answer questions about companies, products, policies, and other dynamic information, providing probabilities and cited evidence. The model supports boolean, choice, and score question types, with related questions sharing context and evidence in a single request. It can be used to enrich CRM records, verify claims, check eligibility, or route leads. The model name is 'parallel/parallel-decisions-fast', requiring no separate Parallel account. Citations include source URLs, optional titles, and excerpts, applicable to the entire request. Applications can act on high-probability answers and route uncertain ones for review. Requests are tracked in AI Gateway logs and spend budgets.

Protein Startups Are Selling Lots Of Snacks And Drinks. Our Sample Tasting Found Most Neither Great Nor Horrible

As we chronicled earlier this week, both startups and established food brands are making it simpler than ever to up protein intake. From sodas to snack bars to ice cream, it’s now possible to get a fix of this coveted macronutrient with foods and drinks more commonly associated with sugar, starch and fat. But, do they taste good? To find out, we hit up a local Target for a selection of protein foods sold by venture-backed startups and subjected them to a taste test. Here are our findings.

Provably Optimal HBT Legalization in Seconds: Certified Optima Reveal Where the Headroom Really Lies

arXiv:2611.14928v1 Announce Type: new Abstract: Hybrid bonding terminal (HBT) legalization assigns every terminal of a face-to-face 3D IC to its own pitch-grid site with minimum total Manhattan displacement. Open3DBench reports that a reinforcement-learning (RL) legalizer reduces displacement by 19-22% relative to a sliding-window (SW) heuristic, but none of its legalizers is publicly released. We ask how much displacement any legalizer can still recover. We build an open benchmark of 21 design/pitch instances from Open3DBench placement data, with reconstructed greedy, SW (ViaLegal-style) and BTAssign-style baselines, and GEO-Exact, an exact solver that runs successive shortest augmenting paths over lazily generated spatial candidate edges. An independent dual certificate proves its assignment optimal on all 21 instances (relative primal-dual gap at most $8.2 \times 10^{-16}$). Against these optima, our SW reconstruction with ViaLegal's default configuration is on average 0.011% above optimal (worst 0.119%), while on the 16 contended instances BTAssign-style, a density-guided heuristic (DGHL) and greedy are 3.0%, 6.9% and 64.5% above. Headroom does appear outside this regime: with 2x2 windows SW reaches 34.1%, and above about 58% utilization, on three of six swept designs, it reaches 21.1% on bp, 27.5% on swerv and 8.9% on jpeg (217 further certified instances, and ten bp points checked against an independent exact solver). If the published 19-22% gain were measured on instances like ours, SW would have to be at least 24-28% above optimal on average, far more than any SW variant we tested at nominal pitch; the RL paper's own comparison with the Hungarian optimum puts its SW 3.05% above optimal on average. GEO-Exact is also about an order of magnitude faster than DGHL and SW in our harness (geometric means 8.1x and 40x; a compiled core against Python baselines on a shared server). Code, optimal values and certificates are released.

Show HN: Deccy – software engineering as decision management

Hi HN,We built deccy out of frustration from working with coding agents.Although we love the speed that AI has brought to our work, we also want to have control over what we build. While it's not impossible to keep up with the changes AI makes, it is significantly more difficult when compared to writing code ourselves.We realised that what we cared about most was not controlling every line of the implementation, but preserving our intent (we especially hated when agents would wrongly guess what we wanted). We iterated through planning and spec-driven methods, but found that both writing our own plans/specifications and reading AI-generated versions was tedious and exhausting.Eventually, we decided to split the specifications up into the decisions that make them up, and get agents to do the boring work of tracking and bookkeeping. deccy is the first realisation of this idea. We're already working on the next iteration, but we wanted to share the mental model we developed.There's a long way to go from here, and we'd really love any feedback :) Comments URL: https://news.ycombinator.com/item?id=50483376 Points: 1 # Comments: 0

Show HN: TempCD – Temporarily CD in a directory with one command

TempCD is just a small thing I made, its only purpose it to eliminate having to enter cd again after only wanting to do one command in a directory. Super simple, just:> tcd ~/example ~/example> command command output >That's it, just returns you back to where you were before without having to type cd again. Comments URL: https://news.ycombinator.com/item?id=50477500 Points: 1 # Comments: 0

Termination Lag: Coding Agents Solve Early but Stop Late

arXiv:2611.14960v1 Announce Type: new Abstract: Coding agents can solve a task well before they stop working on it. We characterize \emph{termination lag}--the number of steps between an agent's first correct patch and the end of its run--across five models, two agent frameworks, and three SWE-bench datasets by recording and testing every distinct repository patch generated during execution. Our analysis reveals that 38--82% of successful runs achieve a correct solution on their initial source-code edit. However, agents continue to expend substantial effort after correct patches, with post-solution steps accounting for a median of 35--37% for Qwen3-Coder and 54--60% for Laguna-XS.2 in successful runs on SWE-bench Verified and Multilingual. Agents spend most of these additional steps verifying their work and exploring the repository. In a small fraction of cases (1--8%), this continued activity leads agents to undo their earlier correct patch, leaving the final solution incorrect.

ToolTurns: Benchmarking LLM Agents in Multi-Turn Dialogue

arXiv:2611.14925v1 Announce Type: new Abstract: Tool-using language models are increasingly deployed in conversational settings, yet most large-scale tool-use benchmarks focus on isolated requests. We introduce ToolTurns, a benchmark of 2,620 problems spanning 1,827 tools and 11,004 API endpoints across 47 categories. Each dialogue begins with an intentionally underspecified request and continues through two context-dependent follow-ups, testing clarification, tool use, and retention of conversational context. We also introduce an open-source user simulator that reveals withheld context through dialogue and generates follow-up requests, with a checker--fixer pipeline for improving its turns. Across nine language models, the best agent resolves 69.6% of individual requests but completes only 39.8% of full dialogues. Under this evaluation protocol, every evaluated model receives lower success scores on follow-up requests than on the initial request, and dialogue-quality rubric scores and task success can rank models very differently. Paired turn-level evaluation shows that the checker--fixer reduces detected dialogue defects, and blind human evaluation favors the revised turns. ToolTurns provides a testbed for evaluating whether agents sustain task success across multi-turn dialogue with tools.