Kappy Prasad
Tech entrepreneur and engineer with 18+ years across networking, cloud, DevOps and security. These days I spend most of my time building with AI: agents that do real work, automations that remove busywork, and the infrastructure that keeps them reliable.
Writing on AI and engineering
AI workforce
An AI workforce, not another chatbot
What the project is, the roles its AI agents take on inside a business, and what changes for the people who work alongside them.
The architecture of an AI workforce
How the project's agents are put together, from role definitions, tools and memory to the coordinator that routes work and the tracing that shows what happened.
Guardrails and approvals for AI agents
How the project keeps agents useful without letting them loose, using least-privilege tools, tiered actions, human approvals, audit trails and evaluations before every change.
Lessons from building an AI workforce
What worked, what didn't and what surprised us while building the project, from narrow roles and better tools to why evaluation turned out to be the real product.
Use cases
Safe AI adoption for engineering teams
Rolling out an enterprise AI coding assistant behind company identity, with guardrails that made the safe path the easy one, then using it to automate the IT team's own manual work.
Security governance and incident response from zero
Standing up a security governance and incident response capability aligned to a national cyber security framework, then proving it through live incidents and exercises, not just audits.
Multi-cloud infrastructure as code
Moving infrastructure across GCP, AWS and Azure into reviewed, pipeline-driven code, and running multi-tenant Kubernetes for multiple customers on the same foundations.
Workforce scheduling SaaS at global scale
Taking a shift management platform from serving individual customers to a global rollout for one very large enterprise, by hardening delivery, automating the path to production and rebuilding support around real KPIs.
Cloud cost optimization
Cutting engineering cost by 20% by making cloud spend visible, owned and reviewed, without trading away reliability or delivery speed.
Network automation on a global backbone
Python tooling that logged into core routers, ran the standard diagnostics and returned a clean summary, taking the repetitive work out of 24×7 operations on a global MPLS/IP backbone.
Clustered hosting platform on AWS
Founding, building and running a clustered hosting platform on AWS for small and medium businesses, with the engineering, the budget and the team all owned end to end.
Engineer, founder, consultant.
I started in networking and infrastructure, moved through software, cloud and security, and founded companies along the way. As a consultant I've worked with more than 100 companies across enterprise, government, healthcare, education and startups. Now I'm putting that foundation to work on AI.
Certifications
AI agents & automation
LLM-powered agents and workflows that take real actions — research, ops, content and code.
Cloud & DevOps
AWS, GCP and Azure; Kubernetes, Terraform and CI/CD built to scale and stay up.
Security
DevSecOps, compliance and risk — shipping fast without leaving the door open.
Let's build something.
Advisory, collaboration, or just comparing notes on AI.