Verified
@JeffHaynes
Georgia, United States
How much access do you give your AI Agent
I’m thinking through how larger engineering teams should safely use coding agents.
Here’s the scenario:
An agent is updating one Java/Spring Boot service inside a larger enterprise platform.
My default instinct is that it should not get full monorepo access.
Provide it with the necessary service, related tests, and contracts it depends on, along with sufficient architectural context to avoid making incorrect assumptions.
However, I would not want to casually read unrelated packages, internal documents, secrets, credentials, or configuration that has nothing to do with the task.
That feels similar to how we already think about production services.
A service should not get access to every database just because it might need one table.
So my question is:
Where would you draw the boundary for coding agents?
Task-scoped access by default?
Full repo access with logging?
Human approval before expanding context?
Different rules for low-risk vs high-risk code?
Curious what others are seeing.
I am a Lead Principal Software Engineer with 25+ years of experience building and modernizing enterprise platforms at scale. My career has had two phases, and the combination is what makes the work I do today possible. The first decade was in professional services, working directly with enterprise clients - gathering complex business requirements, translating them into technical solutions, and learning what enterprise systems actually need to do under real-world load. The next 15 years were at Oracle, where I owned technical vision for the commerce portion of the CPQ platform, leading the shift from legacy architecture to cloud-native, microservices-based systems on OCI. The combination means I do not just build complex systems - I understand deeply why they need to work the way they do. What I bring: • 15 years of bridging legacy enterprise Java systems to cloud-native OCI architecture at enterprise scale - a combination of skills that very few engineers in this market have. • Deep ownership of core platform services (approvals, print, package transformation, layout manager) serving 600+ enterprise customers and 750K licensed users. • Mentorship and technical guidance for distributed teams of 7-10 engineers. • Production reliability and modernization across 2-4 SaaS release cycles per year. Core expertise: Java, Spring Boot, REST/JAX-RS, OCI, Kubernetes, Docker, distributed systems, Oracle CPQ, CI/CD, AI integration, and production troubleshooting. I write about platform modernization, AI integration in enterprise systems, and the bridge between legacy and cloud-native architecture on my site: jefferyhaynes.net. You will find articles, working AI apps, and architecture walkthroughs there. Open to Lead Principal Software Engineer and Staff Software Engineer roles in Atlanta or remote. Reach me at jeff@jefferyhaynes.net.
0
Followers8
FollowingJul 23, 2026
User Since