# Vetro

> Vetro is an AI assurance and governance firm - red-teaming, evaluation, governance and monitoring for LLM apps and AI agents in production. Start with a read-only AI audit.

Vetro is an AI assurance and governance firm based in India. Founder: Prithviraj Chawla. Contact: vetro.team.admin@gmail.com. Full text of every page: https://vetro.co.in/llms-full.txt

## Start here

- [Vetro | AI Security, Accuracy & Governance](https://vetro.co.in/): What Vetro is, the five service lines, how an engagement runs, and the questions people ask first.

## Services

- [Services | Security, Accuracy, Governance & Monitoring - Vetro](https://vetro.co.in/services): The five service lines in full, what each includes, and how they attach to one another.
- [AI Red Teaming & Agent Security Testing | Vetro](https://vetro.co.in/services/security-and-red-teaming): Adversarial testing of what your AI systems can actually reach - prompt injection, tool-call abuse, data exfiltration - scoped by what an agent is able to do rather than by what it was told to do.
- [LLM Evaluation & AI Accuracy Testing | Vetro](https://vetro.co.in/services/accuracy-and-evaluation): Measuring what your system actually gets wrong, on your own data, behind a regression gate that fails the build - rather than a demo that happened to work on the day it was shown.
- [AI Governance Consulting & ISO 42001 Readiness | Vetro](https://vetro.co.in/services/governance-and-policy): Written rules for what your AI systems may do, turned into controls that actually run - approval gates, scoped permissions, retention - plus the inventory and records an auditor or a customer will ask you for.
- [LLM Observability & AI Monitoring | Vetro](https://vetro.co.in/services/monitoring-and-response): Tracing every model call and tool call so you can say what a system did, on what input, and on whose authority - built before the incident, because it cannot be added afterwards.
- [LLM Cost Optimisation & AI Efficiency | Vetro](https://vetro.co.in/services/efficiency-and-cost): Cost and latency measured per outcome rather than per token - model routing, caching, context discipline - and finding the calls that spend money without changing the answer.

## Writing

- [Insights | Practical Guides on AI Security, Auditing and Evaluation](https://vetro.co.in/insights): Long-form guides, each answering one question a buyer actually asks.
- [What Is AI Assurance? A Working Definition, and What It Is Not | Vetro](https://vetro.co.in/insights/what-is-ai-assurance): AI assurance is evidence that an AI system does what it should, cannot do what it must not, and can be shown to have done either. What that means in practice, and how to tell if you have it.
- [Prompt Injection: Why the Fix Is Permissions, Not a Better Prompt | Vetro](https://vetro.co.in/insights/prompt-injection-explained): Prompt injection is not a prompting bug, and it is not patched at the prompt layer. Here is what it actually is, and what genuinely reduces the damage.
- [What an AI Audit Trail Has to Contain to Be Worth Anything | Vetro](https://vetro.co.in/insights/ai-audit-trail): Most AI logging records the answer and throws away everything that produced it. Here is what a trail needs to contain to survive an incident or an audit.
- [Measuring an AI Feature: What to Check Before You Ship It | Vetro](https://vetro.co.in/insights/ai-evaluation-before-you-ship): Twenty prompts that looked right is not a measurement. What an eval set needs to contain, what to measure, and how to stop regressions reaching users.

## Reference and company

- [AI Assurance Glossary | Prompt Injection, Blast Radius, Evals, Tracing - Vetro](https://vetro.co.in/glossary): Short standalone definitions of every term the site uses, each linked to the service line it belongs to.
- [AI Audit | Read-Only AI Risk Diagnostic - Vetro](https://vetro.co.in/ai-audit): The no-fee, read-only diagnostic Vetro runs before any engagement - what it checks, what it produces, and what access it needs.
- [Sample AI Audit Report | What The Audit Delivers - Vetro](https://vetro.co.in/ai-audit/sample-report): A fictional, clearly labelled example of the written report the no-fee audit delivers.
- [About Vetro | Why The Model Is Not The Problem](https://vetro.co.in/about): The firm behind the work - its premise, the four-stage engagement model, its independence and limitations, and founder-led delivery.
- [FAQ | Access, Pricing, Timelines & Limitations - Vetro](https://vetro.co.in/faq): Fifteen direct answers on how Vetro works: access, pricing, timelines, guarantees and exits.
- [Contact Vetro | AI Audit & Discovery Call](https://vetro.co.in/contact): Book a 20-minute discovery call or send the details in writing.
- [Trust & Security | Access, Data Handling & Disclosure - Vetro](https://vetro.co.in/trust): For security and procurement teams: access model, rules of engagement for testing, client data handling, AI tool use, third parties, incident notification and vulnerability disclosure.