Agentic AI Pipeline
Building autonomous agentic AI systems: AI that plans, uses tools, and carries multi-step work through to a finished result, with a person owning the standards and outcomes.
View projectTen years on cloud, DevOps, and security for media, healthtech, and logistics teams. Now I'm building Temika Cyber, and this site is where I write down what I figure out along the way.
Or say hi: LinkedIn [email protected]
I am Co-Founder of Temika Cyber, a cyber security consultancy, and have spent the last decade building and securing cloud infrastructure for companies in media, healthtech, and nationwide logistics. The job is mostly invisible. That is how it should be.
Lately the hardest problems are not human. AI agents, service accounts, and machine identities now outnumber people in the stack, hold more privilege, and answer to no one. I think we should treat them like the identities they are: scoped, credentialed, and auditable. That conviction is most of what I write about here, and what we work on at Temika Cyber. I would rather ship something a team can actually operate than something clever nobody can touch.
A few things I have built, shipped, and learned from.
Building autonomous agentic AI systems: AI that plans, uses tools, and carries multi-step work through to a finished result, with a person owning the standards and outcomes.
View projectTemika Cyber helps Indonesian businesses protect systems and data through assessment, monitoring, and risk handling by a certified security team.
View projectFull-stack software studio where clients track every sprint in a live portal. Code, legal, and growth covered end to end, every deliverable signed off.
View projectHonest notes on cloud, Kubernetes, CI/CD supply chain, and the security of what runs on them.
A signed A2A agent card still lets attackers inject instructions into the description field your orchestrator feeds directly to the LLM router.
Read more: A2A Agent Card Poisoning: JWS Proves Origin, Not IntentThe fix for AI agent identity is two layers: SPIFFE proves what the agent is, OAuth token exchange proves what it may do right now and for whom.
Read more: AI Agent Identity Needs SPIFFE and OAuth, Not One KeyHow to discover, scope, and govern service accounts, API keys, tokens, and AI agents before machine credentials become your breach path.
Read more: Non-Human Identity Governance: Field Tips for 2026I am always glad to talk through a hard infrastructure or security problem, trade notes, or hear what you are building. Reach me by email or on LinkedIn.