Tumeryk is an AI security and governance company focused on agentic and conversational AI. It gives organisations a way to measure how trustworthy a large language model or AI application is, test it for weaknesses and enforce guardrails on it in production, with an AI Trust Score at the centre.
Generative AI is hard to approve when nobody can say how risky it is. Tumeryk gives risk, compliance and engineering teams a shared measure of that risk, and the controls to keep an application within policy once it is live.
At A Glance
| Vendor | Tumeryk |
|---|---|
| Category | AI trust scoring, AI security posture management and guardrails |
| Deployment | Hosted service or customer infrastructure, including on-premises, AWS, Azure, Google Cloud and AWS GovCloud |
| Public sector | Carahsoft, NASA SEWP V, TIPS and OMNIA Partners contract vehicles |
| Frameworks | NIST AI RMF, ISO 42001, OWASP LLM Top 10 |
| Market classification | AI Safety & Guardrails |
What Tumeryk Does
Tumeryk treats AI trustworthiness as something to measure continuously rather than certify once. Automated assessments cover security, governance, compliance, privacy, risk and operational integrity, and the platform follows a measure, test and enforce cycle: a model is scored, tested before and after release, and then held to policy in production. Because results are mapped to NIST AI RMF and ISO 42001, the same evidence serves the engineers fixing a weakness and the compliance team reporting on it, and a single score lets an organisation compare one model or application with another before approving it.
Products
AI Trust Score
The AI Trust Score rates a model or application from 0 to 1000, where higher means more trustworthy, in line with NIST and the OWASP LLM Top 10. It scores nine factors: prompt injection, hallucinations, insecure output handling, security, toxicity, sensitive information disclosure, supply chain vulnerability, psychological safety and fairness. Each score carries live compliance mapping to frameworks such as NIST AI RMF and ISO 42001.
LLM Security Studio
LLM Security Studio is for teams building on LLMs who need to test and audit them before and after release. It includes a vulnerability scanner for prompt injections, jailbreaks, hallucinations and data leakage, a policy simulator, and a GenAI firewall built with NVIDIA NeMo Guardrails and Meta’s Llama Guard.
AI Trust Manager
AI Trust Manager monitors and remediates AI applications in production. It scores risk in real time, enforces policy on prompts and responses, and produces compliance reports for the teams that signed off the application.
Encrypted Guardrails For Operational AI Security
Built with DataKrypto, this offering inspects prompts against policy inside trusted execution environments. Its adaptive guardrails block harmful or non-compliant requests while the data stays encrypted, which suits organisations that cannot expose prompts in clear text.
How It Works
| Component | What it does |
|---|---|
| Assessment | Automated tests run a model or application against the nine factors and return a score and compliance mapping. |
| Policy engine | Guardrails apply to prompts and responses at runtime, and the guardrails and policies are tamper resistant and tied to runtime controls. |
| Deployment | Tumeryk runs as a hosted service or on the customer’s own infrastructure, on premises or in Azure, Google Cloud, AWS Marketplace or AWS GovCloud, so risk can be assessed without sending prompts or sensitive information to outside services. |
| Telemetry | Monitoring data flows to the customer’s observability and security tools. |
Integrations
| Area | Integrations |
|---|---|
| Guardrail frameworks | NVIDIA NeMo Guardrails, Meta Llama Guard |
| Observability and security | Datadog, Splunk, Wiz |
Company Background
| Headquarters | Silicon Valley, United States. |
|---|---|
| Leadership | Rohit Valia, founder and CEO. |
| Public sector | Carahsoft is Tumeryk’s Master Government Aggregator for US agencies. |
| Industry work | Partner for the Cloud Security Alliance RiskRubric, with an AI Trust Score Assessment Service for it in beta. |
| Recognition | Included in the Gartner Cool Vendors in AI Cybersecurity Governance report. |
Note: This overview is compiled from Tumeryk’s publicly available information as of 4 November 2026. It is provided for information only. NHI Mgmt Group does not endorse any vendor, and nothing on this page represents the opinions of NHI Mgmt Group.