News Hub
What actually shipped in agent engineering, pulled from the labs, arXiv and Hacker News.
See who we followBoundaryML built 6 BAML SDKs for 6 programming languages in 1 week with human operators.
Structured outputs in LLMs can create false confidence in results.
Large language models lack true numerical understanding capabilities.
BoundaryML invites developers to create the next coding agent.
Using UUIDs in prompts negatively impacts LLM performance.
BoundaryML shared notes from an advanced prompting workshop.
BoundaryML introduced Lambda, an AI agent system.
Llama API now supports tool use with reasoning capabilities.
BoundaryML demonstrated structured outputs capability with Llama 4.
QwQ 32B model supports structured outputs for constrained generation tasks.
Gemini 2.0 supports structured outputs for deterministic response formatting.
OpenAI o3-mini supports structured outputs.
BoundaryML announced launch week day one event.
BoundaryML proposes new syntax for AI agents.
Deepseek R1 supports structured outputs for deterministic response formatting.
OpenAI O1 model supports structured outputs for constrained generation.
The Agent Architect
One pattern, one tradeoff, one production failure story. A short weekly briefing for people building agentic systems.
Weekly email, one-click unsubscribe. We only use your address to send the briefing.