DEV Community

Brian Hall profile picture

Brian Hall

Co-Founder at Faramesh Labs. I mostly think about all the stuff AI agents shouldn't be allowed to do.. I'm here to write about that, but also probably learn more than I "teach"

Location Boston, MA Joined Joined on  Personal website https://faramesh.dev/

Education

Syracuse University

Work

Co-Founder @ Faramesh Labs

How to stop an AI agent from burning $47,000 in a loop nobody noticed.
brianrhall profile

How to stop an AI agent from burning $47,000 in a loop nobody noticed.

#ai #agents #devops #security
3
Comments 1
4 min read

Want to connect with Brian Hall?

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

Create Account
Already have an account? Sign in
Put a hard stop in front of your CrewAI crew's tool calls
brianrhall profile

Put a hard stop in front of your CrewAI crew's tool calls

#ai #agents #crewai #security
3
Comments 6
5 min read
Don't use an LLM to decide what your AI agent is allowed to do

Soft intelligence vs hard enforcement
brianrhall profile

Don't use an LLM to decide what your AI agent is allowed to do

#ai #agents #security #llm
10
Comments 11
4 min read
Your coding agent will route around your rules. Here's how to actually stop it.
brianrhall profile

Your coding agent will route around your rules. Here's how to actually stop it.

#ai #claude #security #devtools
10
Comments 12
5 min read
How to add policy enforcement to a LangGraph agent (before it does something dumb)
brianrhall profile

How to add policy enforcement to a LangGraph agent (before it does something dumb)

#ai #langgraph #python #security
3
Comments 3
5 min read
Your agent's guardrails are suggestions, not enforcement
brianrhall profile

Your agent's guardrails are suggestions, not enforcement

#ai #claudecode #langchain #security
2
Comments
4 min read
loading...