Agentic engineering, enterprise adoption, evals and reliability, security, agent interfaces, and agentic AI 101. Pick your track and submit to speak at AGNTCon + MCPCon North America 2027, April 28-29 in San Francisco. Deadline: January 18. Submit here >> bit.ly/4tFfCdz
Agentic AI Foundation
Verified account This account is verified. Learn more@AgenticAIFdn
AAIF is a neutral, open foundation ensuring AI capabilities evolve transparently and collaboratively, advancing adoption of leading open source AI projects.
98 Following 10K Followers 1.9K Posts
Latest posts 5
A short AGENTS.md file made a measurable difference in a coding agent's work on the TODO Group website.
Across 18 runs: 36% fewer tokens, 32% less time, and 26% lower estimated costs.
AAIF Ambassador @anajsana95 shares the experiment, method, and limitations.
aaif.io/blog/devex-for…
Across 18 runs: 36% fewer tokens, 32% less time, and 26% lower estimated costs.
AAIF Ambassador @anajsana95 shares the experiment, method, and limitations.
aaif.io/blog/devex-for…
What happens when AI agents can actually learn from experience?
@yaboilyrical of @NousResearch breaks down how Hermes Agent builds reusable skills, retains context, and improves over time instead of starting from scratch with every session.
@yaboilyrical of @NousResearch breaks down how Hermes Agent builds reusable skills, retains context, and improves over time instead of starting from scratch with every session.
0:45
Sheela Zemlin, COO of Vertical AI company @SmartEnergyW shares why she pursued the Model Context Protocol Associate (MCPA) certification designed for AI developers and recently launched by the AAIF, from reading the latest #MCP technical specification to gaining a deeper understanding concepts like stateful and stateless.
For business leaders, that technical knowledge builds confidence, creates a common language with product and engineering teams, and opens the door to becoming more active participants in building AI.
Because anyone can be a builder: bit.ly/4eqKmK9
For business leaders, that technical knowledge builds confidence, creates a common language with product and engineering teams, and opens the door to becoming more active participants in building AI.
Because anyone can be a builder: bit.ly/4eqKmK9
0:23