Martin Keen explains why bigger context windows can make AI agents less reliable. He introduces context rot, the four context failure modes of poisoning, distraction, confusion, and clash, and the context engineering fixes that work by removing information rather than adding it.
SLNG.ai introduces Unmute, an open declarative standard for voice agents, addressing platform lock-in and the limitations of current agent architectures. It proposes a shift from LLM-centric, sequential pipelines to declared execution with structural guarantees, enabling portable, cost-effective, and robust voice agents by breaking down complex interactions into modular, context-aware steps. This approach minimizes token usage, enhances determinism, and improves agent reliability compared to traditional single-prompt designs.
An introduction to Exo, a fully recursive AI agent harnessing a novel three-layer architecture (Executor, Harness, Sandbox) designed for autonomous self-improvement. It delves into how Exo surpasses current agent limitations by allowing the agent to edit its own code and policy at runtime, ensuring protected state and isolated execution, and discusses practical implications and the underlying systems philosophy enabling this paradigm shift.
Once GPT 5.3 Codex Spark achieved 1000 tokens/sec on Cerebras, network latency superseded inference as the bottleneck for agents. This talk details how the Codex harness addresses this and other agentic challenges through innovations like WebSocket mode for stateful context, deferred tools for efficient context construction, robust sandboxing (Seatbelt, Bubblewrap, custom Windows solution), and an auto-review subagent to mitigate approval fatigue while ensuring security. It also covers structured actions via 'apply patch' for file edits, shell tools for system interaction, and sophisticated long-horizon goal management, with most distinct features exposed through the open Responses API.
Lance Martin from Anthropic shares insights into building reliable and secure long-horizon agents with Claude. He details architectural principles like decoupling the 'brain' from the 'hands' for reliability and security, implementing independent verifiers for self-correction, and developing advanced self-learning memory systems akin to human memory's in-band writing and offline 'dreaming' consolidation. The talk concludes with a vision for evolving agent harnesses towards organizational-level, proactive, and multiplayer capabilities.
Jeffrey Lee-Chan argues that the critical component for production AI agent systems is not the model but the "harness"—an orchestration layer providing persistent memory and contextual guardrails. He introduces Open Claw, a conceptual framework that separates agent orchestrator managers (handling task specs and history) from specialized workers (executing implementation details). This architecture enables deterministic teamwork, parallelization, and unbiased outcomes, supported by examples like AI RPGs and multi-AI analysis, alongside tools like tmux for enhanced developer productivity and careful token management strategies.