Introducing TernaryPhysics Ops

Today we're launching TernaryPhysics Ops — a new way to investigate and manage your infrastructure using AI agents that live directly on your resources.

The Problem

Infrastructure investigation is painful. When something breaks at 3am, you're SSH-ing into machines, grepping through logs, cross-referencing metrics, and trying to build a mental model of what went wrong. Even with modern observability tools, you're still the one connecting the dots.

Existing solutions either require sending your data to external services (security risk) or running heavy analysis pipelines (complex setup). Most teams end up with dashboards full of metrics but no actual insight into what's happening.

Our Approach: Drop Agents

We took a different approach. Instead of aggregating your data somewhere else, we bring the AI to your data. You "drop" lightweight agents directly onto your resources — Kubernetes clusters, VMs, databases, message queues, anything.

$ ssh my-server
$ pip install tp-ops
$ tp-ops drop

Detecting resource type...
✓ Detected: PostgreSQL 15.2
✓ Agent type: postgres-agent

Agent dropped successfully!

Each agent runs two AI models: the TNN™ (Ternary Neural Network) for continuous anomaly detection, and TernaryPhysics-7B for conversational reasoning when you need it. Both models run locally — your data never leaves your environment.

TNN™ and TNN Mesh™ are Patent Pending technologies of TernaryPhysics LLC.

The Mesh Effect

Here's where it gets interesting. When you drop multiple agents, they automatically discover each other and form a mesh network. When you ask one agent a question, it can query other agents to trace causality across your entire infrastructure.

The more agents you drop, the smarter the mesh becomes. An API latency issue might be caused by a database connection pool exhaustion, which was caused by a recent deployment that changed an environment variable. The mesh traces this causality automatically.

Talk to Your Infrastructure

Once agents are deployed, you can have natural language conversations about your infrastructure:

$ tp-ops ask prod-cluster

prod-cluster > Why is the API slow?

Investigating across mesh...

→ k8s-agent: payment-api latency 3x baseline
  └─ Querying postgres-agent...

→ postgres-agent: Connection pool at 147/150
  └─ Querying k8s-agent for recent deploys...

→ k8s-agent: Deploy 2 hours ago changed POOL_SIZE env var

Root cause: Deploy removed POOL_SIZE config, defaulting to 150.
Connection pool exhausted under normal load.

Fix: Restore POOL_SIZE=50 in payment-api deployment.
Apply fix? [yes/no]

The agent doesn't just find problems — it recommends specific fixes based on what it learns from your environment. And because it runs locally, it has access to everything: logs, metrics, environment variables, deployed configs, process lists, network state.

Pricing: Pay Per GB

We charge $0.50 per GB of data processed by agents. No per-agent fees, no seat licenses, no enterprise minimums. Drop as many agents as you want — you only pay for what you use.

The only data we see is billing metadata: how many GB each agent processed. We never see your actual infrastructure data, logs, or metrics. Everything is processed locally.

Get Started

Installation takes 60 seconds:

$ pip install tp-ops
$ tp-ops drop

We currently support 220+ resource types including Kubernetes, PostgreSQL, MySQL, Redis, MongoDB, Elasticsearch, VMs, API gateways, message queues, and more. Check out our agent catalog for the full list.

We're excited to see what you build with TernaryPhysics Ops. Questions? Email us at ops@ternaryphysics.com.