5/5 Ensemble: Run multiple distinct agent configs in parallel → select the best trajectory. Leverages success of the portfolio vs single variants (see also: @/karpathy’s LLM Council). Can outperform best-of-N on accuracy but at lower cost & latency. Here’s how each method did
AGENTS
-
Best-of-N : optimisation parallèle d’agents IA
By
–
4/5 Best-of-N: Run one agent config N times in parallel → select the best trajectory. Leverages LLMs’ non-determinism – but hinges on a good eval mechanism (we use an LLM-as-a-Judge). Can increase accuracy without linearly increasing latency – but parallel runs can add up $$.
-
Agent Scaffolding: Building the Infrastructure
By
–
3/5 Agent scaffolding: Adjust prompts, integrate tools/skills, set execution policies, retries, memory, etc. → create the operational framework. Useful for long-horizon or multi-step tasks.
-
4 agent optimization methods tested on BrowseComp-Plus
By
–
1/5 We’re seeing 4 common agent optimization methods for hitting the right accuracy-cost or accuracy-latency tradeoff. We tried them all out on BrowseComp-Plus (results at end of the thread).
-
AI voice use cases for podcasts, training, and storytelling
By
–
Think about the use cases: → AI podcasts with distinct hosts
→ Training simulations with realistic dialogue
→ Storytelling with emotional range
→ Assistants that sound more natural
→ Multi-speaker conversations where every voice has a personality This is not just -
AI Agent Must Cut Tokens via MCP
By
–
Your AI agent must connect to its MCP server and follow the workflow to significantly reduce token usage.
-
Monitoring AI agent token usage and performance
By
–
Thanks. Can you conform you’re on the latest version? There should be a section right below the progress bars that gives you a detailed breakdown of the specific agents, skills, plugins, and usage patterns that caused outsized token usage
-
Glean platform uses AI Agents for unified enterprise information retrieval
By
–
Glean Prompt Library! @glean Glean’s platform seamlessly integrates with a wide array of enterprise systems such as email, intranet, cloud storage, and database by providing a powerful, unified search interface with AI Agents that effortlessly retrieves information from all
-
Codex uses Peekaboo to get Telegram token from BotFather
By
–
Codex was debugging a Telegram issue and needed a new token, so it used Peekaboo to open the Telegram Mac app, talked to botfather and just did it. Computer Use is amazing. https://
peekaboo.sh -
Man builds laser drone using Claude Code
By
–
Man builds a drone that tracks targets with a laser using Claude Code pic.twitter.com/ZeTSHZZrXC
— AI Breakfast (@AiBreakfast) 13 mai 2026Man constructs a drone that tracks targets with a laser, powered by Claude Code