Tag
Task Model Induction (TMI) is a method that induces structured task models from raw computer-use traces, improving agent performance by 30% and enabling auditable, reusable skills for real-world workflows.
Google announces the integration of Antigravity AI agents into Gemini Enterprise subscriptions and new IDE extensions to enhance agentic workflows for developers.
Gartner predicts that AI inference costs per agentic workflow will increase more than fivefold by 2028, driven by efficiency gains that enable more powerful models and applications, paradoxically raising overall costs.
A hosted MCP server provides a unified API for agent-driven social media management, enabling automated content workflows with human approval gates and platform-specific handling.
This article explains the 'Gauntlet Looping' technique for using AI agents to generate high-quality outputs, demonstrated by building a Call of Duty-style game from a single prompt with Claude Opus 5, and provides a step-by-step guide for implementation.
Mark Cuban advises that learning AI and agentic workflows is key to capitalizing on the AI gold rush, as businesses will need help implementing AI solutions.
Andrew Ng has released a free 1-hour course on becoming an AI agentic engineer, covering AI agent basics, agentic workflows, design patterns, and practical tips for building AI agents.
This paper introduces Matrix, a deterministic causal-state layer for agentic workflows that records authority and fact dependencies, verifies completion evidence, and selectively invalidates stale work to ensure governed execution and auditability.
Koray Kavukcuoglu announces the launch of Gemini 3.7 Flash, a new coding and agentic-workflow model offered at half the original cost of 3.6 Flash. The team highlights a CodeArena score of 1588 and a commitment to improving the coding experience.
Cursor Origin is a new version control and infrastructure product tailored for agentic workflows, featuring high-speed cloning, automatic merge conflict resolution, and one-click GitHub import, rolling out to Cursor users now.
FlowScout is a framework that automatically generates tool-integrated agentic workflows from historical task-solving records, using Monte Carlo tree search guided by execution feedback. Experiments show it improves tool invocation correctness and execution score over baselines.
Jerry Liu shares thoughts on how forward-deployed engineer (FDE) work will shift toward defining goals, environments, and evals while automated optimization processes handle tactical implementation.
Andrew Ng's 2024 prediction that agentic workflows with weaker models can outperform stronger standalone models is validated by Anthropic, where an orchestrator coordinating cheaper subagents beat Claude Opus by 90.2%. The post highlights Claude Code subagents as the practical implementation.
A practitioner argues that autonomous AI agents are unreliable in production, advocating for constrained agentic workflows with human-in-the-loop triggers instead of full autonomy.
Not Diamond announces Not Diamond Code, an intelligent model router for coding agents that selects the best model and reasoning effort per step, claiming 20%+ cost savings and Pareto-optimal performance on coding benchmarks.
This paper introduces MetaRoute-Bench, an open benchmark for evaluating meta-decision policies in agentic workflows, comparing routing policies on success, cost, and latency under a shared execution model.
This paper introduces an executable benchmark for learning compositional meta-routing in agentic workflows, where a budget-aware controller composes operations like retrieval, code execution, and verification. The learned policy outperforms static workflows on held-out tests but shows lower lexical generalization on challenge splits.
The Aspire team describes how they used GitHub Agentic Workflows to automatically generate cross-repo documentation pull requests for merged features, achieving 82/82 docs PRs merged at a median of 44.8 hours.
MCP has released a new specification version 2026-07-28 that shifts from a bidirectional stateful protocol to a stateless request/response core, improving scalability and reliability. The update includes self-describing requests, server discovery, caching hints, and authorization hardening, with updated SDKs in TypeScript, Python, Go, and C#.
HeraSys is a collaborative LLM serving system that optimizes end-to-end performance of concurrent workflows by eliminating cross-workflow computational redundancy and using load-aware joint scheduling, achieving up to 2.17x P99 latency reduction and 1.85x throughput improvement.