Maya for research and education

A core promise of Maya is the agentic networking substrate that emits telemetry. We ship intelligent defaults to observe and enforce, and because the substrate is emit-first, you are never locked into our judgment: route the telemetry into the tools you already run, Datadog, Grafana, Splunk, or the LangSmith-style tooling your agent stack uses, or bring your own judge, an LLM, Open Policy Agent, or your SIEM, and program Maya yourself.

That is exactly why we believe labs and universities are an excellent environment to help build this substrate, particularly around agentic network behavior. Researchers need ground truth about what agents actually do on the wire. Maya produces it without touching their code.

A measurement instrument for the agentic era

Maya identifies each agent from the host and observes its traffic payload-blind, never terminating TLS, never decrypting, with no SDK, no code change, and no framework instrumentation. One declaration line per agent, and Maya emits a clean, structured record of what that agent reaches, who it delegates to, and where it strays from what you declared. The telemetry exports as OTLP or JSON for your own notebooks and analysis.

Three ways labs put that to work:

Measure agentic systems. Ground truth for multi-agent runs: per-agent flows, provider calls, message rates, and the agent-to-agent delegation graph across hosts, without instrumenting a single framework.

Study agent safety and security. Declare an agent's authorized envelope, then watch Maya flag when observed behavior contradicts it. That signal is the basis for studying misbehavior, containment, and agents that hold a valid credential yet act out of role.

Teach courses and labs. Stand up a reproducible multi-agent environment students can deploy and reason about, with delegation graphs, deviations, and containment as hands-on, repeatable exercises.

Design partners, not customers

Maya Research Edition is free of cost for research labs and universities. We partner with a small number of public and private research labs, universities, and educational programs. You get the full substrate under a free educational license, a step-by-step runbook, and first-class, hands-on support, because we want to learn from how you push it.

Getting started is deliberately simple:

1. Discovery. A short call to map your host inventory and what you want to study or teach.
2. Provision. We provide two binary packages built for your environment, a host-side package for each agent host and the Warp appliance package that governs their traffic, plus recommended compute specs and the runbook. You deploy on your own lab or cloud infrastructure; your data never leaves your environment.
3. Declare. Add one line to each agent's existing deployment config to name it. No code changes.
4. Observe and experiment. Read per-agent telemetry, map delegation graphs, flag deviations, plug in your own judge.

In return we ask for candid feedback on what works and what's rough, a chance to learn together from real multi-agent deployments, and openness to collaboration, case studies or citations, only if and how you're comfortable.

Where Maya is today

Maya is in active beta. Per-agent observation, delegation-graph mapping, and declared-vs-observed deviation telemetry are what we lead with; programmable enforcement is available and evolving. We're direct about what is proven versus in progress, and a design partnership is exactly where that conversation belongs.

Bring wire-truth to your agent research

Download the Maya Research Edition brochure to share with your lab or department, and tell us about what you're building at research@mayagentic.com. We'll take it from there.