Rebuilding the web for agents — Liad Yosef, MCP Apps
The agentic web is a shift, not a web of agents. We must break interfaces into atoms and let an assistant compose them, moving beyond the need for traditional browser tabs.
Read articleChannel
AI Engineer
Digests from AI Engineer
The Search Engine for the Agentic Web — Will Bryk, Exa
Exa provides a specialized search engine for AI agents, predicting that machine searches will exceed human searches by a thousandfold by 2026. It addresses the limitations of traditional recommendation-based search.
Read article
The unreasonable effectiveness of BM25 for agentic search — Jo Kristian Bergum, Hornet.dev
BM25, a 30-year-old lexical scoring function, is making a comeback for Agentic Search. The change is not in the function, but in the user: the LLM agent, which can formulate highly specific queries and execute complex search trajectories.
Read article
Act, Confirm, or Stop? Smarter behavior for AI assistants, wearables & robots — Amit Desai, Roku
Improving voice AI requires optimizing system behavior under uncertainty, not just accuracy. By minimizing the Outcome User Cost Heuristic (OUCH), developers can significantly boost UX even if the core recognition accuracy remains constant.
Read article
"My name is... my name is...": A Linguistic Map for Voice Agents — Midam Kim, ServiceNow
Voice AI failures are not random bugs, but structural issues. Kim introduces a linguistic framework mapping communication to two channels (listening/speaking) and four interdependent levels: sounds, words, interaction, and mental model.
Read article
Realtime Voice Agents with Frontier Intelligence — Bohan Li, EliseAI
Learn how to build real-time voice agents with Frontier Intelligence. We detail a cascaded architecture using speculative transcription, background tool calling, and prefix caching to eliminate latency.
Read article
5 Voice Agent Failure Modes You'll Hit in Week One — Venky B, Plivo
Voice AI agents often fail in production due to latency, brittle data collection, and poor LLM-to-TTS handoffs. The talk outlines solutions using structured data modeling and normalization layers.
Read article
I Monitored Crime Audio. Voice Agents Scare Me More. — Sumanyu Sharma, Hamming AI
Voice agents are scaling fast, but reliability is the number one problem. The speaker outlines a continuous monitoring loop, emphasizing that centralized AI systems require deep adversarial testing to prevent massive failures.
Read article
Tolan: Voice-First AI Companion — Paula Dozsa, Tolan
Building a voice-first AI companion requires designing for 'conversational volatility.' The team uses tiered model routing based on emotional stakes and rebuilds context per turn, treating memory as a retrieval system.
Read article
Your Voice Agent is Just a Walkie Talkie — Neil Zeghidour, Gradium
The talk details the evolution of voice agents, arguing that achieving full-duplex, natural conversation requires a hybrid architecture: coupling a small S2S interface with a powerful background text LLM.
Read article
Voice Agents Can Just Do Things — Charlie Guo, OpenAI
Voice agents don't have to talk back. The key is understanding the three modes (Speech-to-Action, Event-to-Speech) and exposing your app's verbs as callable tools.
Read article
Speech-to-Speech Model Research at Google DeepMind — Valeria Wu Fon & Tom Ouyang, Google DeepMind
Speech-to-Speech models are revolutionizing voice agents by moving beyond simple ASR/TTS chains. By leveraging natively multimodal pre-training, these models achieve a trifecta of conversational fluency, intelligence, and multimodality.
Read article
Agents Without Code: Skills, YAML, and Filesystems Replaced Python — Philipp Schmid, Google DeepMind
Agents are evolving from complex Python loops to declarative, file-based definitions. The new architecture uses a secure sandbox and server-side state management, allowing developers to focus on instructions and capabilities rather than orchestration code.
Read article
How We Solved Agent Building — Andrew Qu, Vercel
Andrew Qu details the evolution of agent building, moving from manual data queries to the file system unlock. He introduces Eve, a framework bringing Next.js conventions to agents, making it easier to build powerful, context-rich applications.
Read article
No Memory, No Harness: Why the Database Is the Last Line of Defense — Kay Malcolm, Oracle
AI agents are powerful, but they are incomplete without a robust memory system. The database must serve as the central nervous system, unifying relational, graph, and vector data to ensure a single source of truth for enterprise context.
Read article
We let an AI agent execute Bash and lived to talk about it — Sarah Sanders, PostHog
Building agents is like shipping a 'malware starter pack.' Security requires deterministic layers, scanning the entire content supply chain, and ensuring the LLM acts only as an adviser, never an enforcer.
Read article
Every step you take, every call you make: the reliable agent stack — Giselle van Dongen, Restate
Restate provides a durable foundation for building resilient agentic systems. It handles complex state management, recovery logic, and session isolation, allowing agents to survive crashes and long waits.
Read article
Harness Engineering: Building the Production Cage for Powerful Domain Agents — Mike Chambers, AWS
Learn about Harness Engineering: the architectural pattern for building scalable, production-grade AI agents. We discuss separating memory, tools, and runtime components for multi-tenant deployment.
Read article
Tokens Should Have Jobs — Katelyn Lesse & Angela Jiang, Anthropic
Anthropic proposes that tokens should have jobs, not just be spent on execution. By implementing strategies like Advising, Grading, and Dreaming, agents can achieve superior reliability and efficiency within a fixed budget.
Read article
Building ambitious software — Jonathan Kelley, Dioxus Labs & Cognition
Jonathan Kelley discusses building ambitious, cross-platform software with Rust and the impact of AI coding agents. He argues that while code is now cheap, architectural quality is paramount for modern build systems.
Read article