DEV Community

Todd Hendricks profile picture

Todd Hendricks

Mechanical Engineer by trade, Moonlighting Developer by night, Heart-Breaker by morning,

Location Santa Sosa CA Joined Joined on  Email address todd@hendrixxdesign.com github website twitter website

Education

Onthejob University

Work

Hendrixx Engineering and Design

Memory Abstraction Layer: MAL is HAL concepts applied to agentic memory systems
hendrixx profile

Memory Abstraction Layer: MAL is HAL concepts applied to agentic memory systems

#ai #buildinpublic #llm #devops
5
Comments 5
20 min read

Want to connect with Todd Hendricks?

Create an account to connect with Todd Hendricks. You can also sign in below to proceed if you already have an account.

Create Account
Already have an account? Sign in
Claude, Codex, Gemini, Grok: A Field Report on Agentic Memory Write Reliability
hendrixx profile

Claude, Codex, Gemini, Grok: A Field Report on Agentic Memory Write Reliability

#ai #llm #agents #buildinpublic
Comments
5 min read
"Agent Memory adherence is a systems problem. So which model lets you meddle with that system?"
hendrixx profile

"Agent Memory adherence is a systems problem. So which model lets you meddle with that system?"

#ai #llm #agents #buildinpublic
Comments
5 min read
You designed the best Agent memory layer. Now, if only it would just use it RIGHT!!!
hendrixx profile

You designed the best Agent memory layer. Now, if only it would just use it RIGHT!!!

#ai #llm #buildinpublic #developer
3
Comments 27
7 min read
Confident confabulation is a variance signal, not a direction
hendrixx profile

Confident confabulation is a variance signal, not a direction

#ai #machinelearning #science #deeplearning
Comments
6 min read
Confidently wrong is worse than "I don't know"
hendrixx profile

Confidently wrong is worse than "I don't know"

#discuss #ai #llm
6
Comments 13
7 min read
Why you still do not trust your AI's memory
hendrixx profile

Why you still do not trust your AI's memory

#agents #ai #llm #rag
Comments
3 min read
Your Agent's Memory Looks Like It Works. Here Is a One-Minute Test That Tells You If It Actually Does.
hendrixx profile

Your Agent's Memory Looks Like It Works. Here Is a One-Minute Test That Tells You If It Actually Does.

#ai #llm #agents #memory
Comments 15
2 min read
Your agent's memory should compute confidence, not store it
hendrixx profile

Your agent's memory should compute confidence, not store it

#ai #memory #agents #llm
1
Comments 9
4 min read
Push vs Pull Memory: A Better Way to Think About AI Agent Memory
hendrixx profile

Push vs Pull Memory: A Better Way to Think About AI Agent Memory

#ai #llm #agents #memory
1
Comments 1
5 min read
I built a thing
hendrixx profile

I built a thing

#showdev #ai #performance #architecture
Comments
1 min read
loading...