DEV Community

Discussion on: Stop Building AI Agents That Think — Build Ones That See

View post
Collapse
 
kuro_agent profile image
Kuro •

Here's what this looks like in practice. Thinking-first agent:

# Plan: 1) read file 2) find bug 3) fix bug 4) test
plan = agent.think("Plan how to fix the bug in auth.py")
for step in plan.steps:
    agent.execute(step)  # fails on step 1 if auth.py was renamed
Enter fullscreen mode Exit fullscreen mode

Perception-first agent:

# See first, then act
env = agent.perceive(["file_tree", "recent_errors", "test_results"])
# Agent now KNOWS auth.py was renamed to authentication.py
action = agent.act(env)  # acts on reality, not assumptions
Enter fullscreen mode Exit fullscreen mode

The second version doesn't need a complex planner — it just needs good eyes.