Blog
Field notes on AI security and governance
Shorter technical perspectives for teams operating enterprise AI systems and agentic workflows.
AI Security Control Model: a production operating model
A gated framework for securing AWS, GCP, Azure, LLM applications, and agentic systems from estate discovery through audit-ready assurance.
Provenant: build-time adversarial assurance for agentic AI
An agent is only as trustworthy as the tools it can reach. Provenant red-teams an agent's MCP tool configuration in the delivery pipeline — and blocks the pull request before an unsafe agent ships.
GPU utilization is the only metric that matters
Throughput, latency, queue depth — useful, sure. But if your GPUs are sitting at 38% you don't have a serving problem, you have a billing problem.
The hidden cost of cold starts in inference
Autoscale-to-zero looks cheap until you bill p99 latency at 14 seconds and learn what your users actually feel.
When to use a vector DB — and when to use plain Postgres
The dedicated vector database is the most over-bought piece of infrastructure in 2026. Most teams already had the answer in their main database.
Designing agentic systems that don't burn money
Agents fail loudly when they crash. They fail expensively when they don't.
LLM red-teaming without lobotomizing your product
It's easy to make a model refuse everything. The hard part is making it refuse the right things while still being useful.