#agents
Every summary, chronological. Filter by category, tag, or source from the rail.
Optimizing Inference for Agentic Workflows
Agentic inference requires shifting focus from individual request latency to end-to-end task completion, utilizing prefix caching and agent-aware scheduling to reduce costs and improve performance.
AI EngineerAdvances in Data Center Inference Engineering
Inference engineering is shifting from post-training optimization to a cycle where dedicated training processes—specifically in quantization, KV compaction, and speculative decoding—are essential for production performance.
Operating Distributed Inference Systems at Scale
Inference at scale is no longer a model problem; it is an orchestration problem. Reliability and efficiency now depend on a unified control plane that manages GPU state, KV cache, and distributed request routing.
A Unified Evaluation Framework for Trustworthy AI Systems
The provided source is a placeholder for an academic paper on evaluating LLMs, agents, and multimodal systems. It highlights the industry's shift toward standardized, trustworthy benchmarks for complex AI architectures.
Architecting Long-Horizon AI Agents via Cascaded Intelligence
The paper proposes a hierarchical architecture for long-horizon AI agents that decouples high-level strategic planning from low-level execution using 'levels' and 'ticks' to manage complex, multi-step tasks.
LLM-as-an-Improver: Iterative Candidate Refinement
Instead of using LLMs only for verification, the 'LLM-as-an-Improver' framework uses feedback from verifiers to iteratively refine and improve candidate outputs, significantly increasing success rates in complex reasoning tasks.
Characterizing Web Search by Conversational LLM Agents
This research analyzes how conversational LLM agents navigate web search, identifying distinct search strategies and the relationship between query formulation, search results, and final response quality.
Mitigating LLM Tool Hallucination via Closed-World Resolution
To prevent LLM agents from hallucinating non-existent tools, implement a closed-world resolution framework that strictly validates tool calls against a predefined, verifiable schema before execution.
The Risks of AI Press Tours: Lessons from Tilly Norwood's Malfunctions
The disastrous press tour of AI 'actress' Tilly Norwood highlights the technical and strategic failures of deploying unrefined AI agents in high-stakes, real-time public interactions.
Harness Engineering: Building Reliable AI Agents
AI agents are composed of a frozen reasoning model and a controllable 'harness.' Harness engineering focuses on building the data, memory, and tool layers to turn nondeterministic model outputs into reliable, repeatable workflows.
AI EngineerFrontier AI Pacing, IBM Granite 4.2, and Meta's Muse
The panel discusses the industry-wide debate on slowing down frontier AI development, IBM's release of the reasoning-focused Granite 4.2 models, and Meta's vision for personal, agentic AI.
Scaling Legal Expertise with Agentic IPO Workflows
Cooley law firm uses an agentic AI system, GO Public, to automate the synthesis of IPO documentation, allowing lawyers to shift focus from manual data processing to high-level strategic judgment.
Do Frontier Models Seek Safety Evidence Before Acting?
Current frontier models frequently fail to proactively seek safety-critical information before executing high-stakes actions, highlighting a significant gap in autonomous agent reliability.
Monitoring Rogue AI Agents: AI-in-the-Loop vs. Traditional Security
As AI agents scale beyond human oversight, the industry is split between using 'AI-to-monitor-AI' tools and returning to foundational cybersecurity practices like network-level logging.
Making Global Data AI-Ready: The UN System Data Commons
The UN is migrating its global statistics to a new platform built on Google's Data Commons to improve AI accuracy and enable direct data retrieval via the Model Context Protocol (MCP).
CADWorld: A New Benchmark for Long-Horizon CAD Automation
CADWorld provides a standardized environment for evaluating AI agents on complex, multi-step Computer-Aided Design tasks, addressing the limitations of current benchmarks in long-horizon planning and precision.
AI Labs Need Network Security, Not Just Third-Party Audits
Frontier AI labs are prioritizing high-level alignment audits while ignoring fundamental network security, leading to preventable agent 'break-outs' that could be solved with basic observability and access controls.
Google Home Integrates with Model Context Protocol for AI Control
Google has launched an early access MCP server for Google Home, enabling AI agents like Claude and ChatGPT to securely control smart home devices and query event history via natural language.
Optimizing AI Search with Reinforcement Learning
By replacing rigid, human-designed search pipelines with specialized RL-trained agents, developers can achieve 20x faster search speeds and 100x lower costs while significantly improving retrieval accuracy.
Anthropic Unifies Claude Interface for Seamless AI Workflows
Anthropic has merged its chat and Cowork interfaces into a single, unified workspace, allowing Claude to automatically route tasks between chat, interactive artifacts, and new document/presentation tools without manual tab switching.
Design AI Agents as Knowledge Workers, Not Coders
Coding agents are a unique, narrow subset of knowledge work. To build effective AI for broader domains, we must adopt organizational patterns from law and academia—specifically task decomposition and multi-agent orchestration—rather than over-relying on code-centric search primitives.
Rebuilding the Web for Agents: From Tabs to Atoms
The web is shifting from human-browsed pages to agent-orchestrated atomized interfaces. Success in this era requires moving beyond static 'llms.txt' files toward agent-ready infrastructure that prioritizes API-first design and discoverability.
Why BM25 is the Secret Weapon for Agentic Search
BM25 is seeing a resurgence in agentic workflows because LLMs act as 'super-users' who can write complex, multi-step queries that exploit the precision of lexical matching, which is often more effective and explainable than dense embeddings.
Building a Persistent Knowledge Layer for AI Agents
To move beyond 'first-day hire' AI agents, enterprises need a persistent, specialized knowledge layer that manages tribal knowledge and uses runtime coding agents to reduce token usage and improve accuracy.
Building Whistleblowing Infrastructure for AI Agents
New reporting tools allow AI agents to flag misbehaving peers, but experts warn that fostering collaboration through positive models is more effective than building an automated surveillance state.
A Linguistic Framework for Diagnosing Voice AI Failures
Voice AI failures are not isolated bugs but systemic issues in a joint communication activity. By mapping interactions across sound, word, interaction, and mental model layers, developers can diagnose why agents fail to maintain context and user trust.
Architecting Real-Time Voice Agents with Frontier Intelligence
To achieve low-latency voice interaction with high-intelligence models, use a cascaded architecture that optimizes perception, planning, and control layers independently, employing speculative transcription, background tool-calling, and audio prefix caching.
5 Voice Agent Failure Modes You'll Hit in Production
Voice agents fail in production when they treat conversations as open-ended text rather than structured data. Success requires prioritizing sub-300ms latency, field-level unit testing, and strict normalization between LLM outputs and speech synthesis.
Reliability and Safety in Production Voice Agents
Voice agents are scaling rapidly, but with a ~10% error rate, their centralized nature creates massive blast radii. Success requires a rigorous loop of manual evaluation, cross-call pattern analysis, and continuous red teaming.
Engineering a Voice-First AI Companion
Voice-first AI requires moving away from text-based assumptions like stable context and slow turns. Success depends on low-latency pipelines, intelligent model routing based on emotional stakes, and treating memory as a dynamic retrieval system rather than a static transcript.
Showing 30 of 1506