Mindset AI
6.8/10Mindset AI is a platform that mid-market and enterprise B2B SaaS companies use to build, embed, govern, and improve customer-facing AI agents inside their own products.
Data last updated November 2026 · Our methodology
Overview
Mindset AI is a London-based platform for software companies that want to ship conversational AI agents inside their own products. Product managers, solutions engineers, and developers work in a shared environment. They design agents, connect them to the company's APIs and third-party apps through named operations, then embed the agents in a web page or a custom UI with an SDK. Agents can also be started from a server, a schedule, or a webhook, and the platform itself can be driven from AI coding tools through an MCP server. One agent can run in different versions for each customer or pricing tier, with different tools, knowledge, and limits, which lets SaaS vendors package AI as an upsell. Changes are versioned in a Sandbox and activated separately for Production, and any version can be rolled back. Operators review run history, failed runs, and costs, and can A/B test cheaper models. Governance is a strong point: a tamper-evident audit chain, optional masking of personal data before it reaches the model, human approval for actions that write to other systems, and evidence packs mapped to frameworks such as SOC 2, GDPR, HIPAA, and the EU AI Act. Customers bring their own Anthropic, OpenAI, or AWS Bedrock keys, and the platform can be self-hosted in the customer's own AWS account. Many early customers are learning and content businesses, such as Mindtools Kineo, Hult Ashridge, and Fuse Universal. Pricing is not published, so buyers have to talk to sales. Public reviews are few, and several safety defaults, including personal data masking and action approval, must be switched on manually before go-live.
Our Rating
Who Mindset AI is best for
Best software for SaaS CompaniesGood fit
The platform is built specifically for B2B SaaS vendors embedding agents in their products, with per-customer and per-tier agent versions.
Product, design, and solutions teams can build and tune agents without waiting on an engineering backlog.
Many named customers are learning and development providers that use agents to deliver and monetise their content.
The vendor and directory listings name mid-market SaaS vendors as a core target.
An SDK, API, webhooks, an MCP server, and migration from LangChain give developers hooks, though the focus is collaborative building rather than code-first work.
Audit chains, compliance evidence packs, data residency, and AWS self-hosting suit enterprise buyers, though the company itself is small.
Less suited
Quote-only pricing, with directory figures in the hundreds to thousands per month, and its SaaS-embedding focus make it a poor fit for small firms.
Pros & Cons
Pros
- Per-customer and per-tier agent versions let SaaS vendors sell AI features as paid upgrades without building separate agents.
- Governance goes beyond most builders: tamper-evident audit chain, personal data masking, human approval of write actions, and evidence packs for SOC 2, GDPR, HIPAA, and the EU AI Act.
- Self-hosting in your own AWS account via CloudFormation has no call-home or licence kill switch.
- Bring-your-own-key support for Anthropic, OpenAI, and AWS Bedrock, with model pinning per agent and A/B testing of cheaper models.
- Capterra reviewers rate customer service 5.0 and praise the dedicated customer success contact and the team's responsiveness.
Cons
- No public pricing, and third-party listings disagree widely, from about $652 per month to £30,000 flat.
- Safety defaults need attention: personal data masking is off and automatic approval of new write actions is on until an admin changes them.
- There is no spending cap; self-hosted deployments only get budget alerts at 85% and 100%.
- Docs say Anthropic and OpenAI calls are not kept in one region, and knowledge embeddings always go through OpenAI.
- Few public reviews (about 15 on Capterra, one on G2), with occasional complaints of slow responses and features still on the roadmap.
Key Features
Mindset AI Pricing
Custom
- Quote-based pricing through sales
- Sandbox and Production environments
- Audit record and compliance evidence packs
- Bring your own Anthropic, OpenAI, or AWS Bedrock key
- Hosted service or self-hosting in your own AWS account
Frequently Asked Questions
Mindset AI does not publish prices, so you need to contact sales for a quote. Directory listings give conflicting figures, such as a starting price of $652.40 per month on SourceForge and a £30,000 flat rate on GetApp UK, and the vendor has not confirmed either.
Yes. The docs describe deploying a single container image into your own AWS account with a CloudFormation template, using ECS Fargate, PostgreSQL, S3, and Secrets Manager. You choose when to upgrade and you are responsible for uptime. The marketing site also mentions GCP and Azure, but the self-hosting guide covers only AWS.
Your organisation sets a key for Anthropic, OpenAI, or AWS Bedrock, and you can pin a model to each agent. Knowledge embeddings use OpenAI. Only on Bedrock do calls stay in the region you choose.
Every action and model call is written to a tamper-evident audit chain, with retention adjustable from 6 to 120 months. An Obligations view maps records to EU AI Act, GDPR, SOC 2, NIST 800-53 AU, ISO 42001, HIPAA, and SEC 17a-4, and can export evidence packs. These are evidence tools, and the docs do not claim a certified legal archive.
You can embed an agent on a page, use your own UI through the SDK, or call agents through the API, a schedule, or a webhook. The platform can also be managed from AI clients through its MCP server.
The vendor says existing agents from frameworks like LangChain can be mapped onto Mindset, carrying over prompts, APIs, MCP tools, sub-agents, and evals.
Capterra lists a free trial and no free version, but the Mindset site sends prospects to a contact form rather than a self-serve signup.
Similar Tools
See all alternatives
LangChain offers open-source frameworks (LangChain, LangGraph, Deep Agents) plus the paid LangSmith platform. Developers use them to build, evaluate, deploy and monitor LLM agents.
Open-source platform for building LLM apps, agents, RAG knowledge pipelines and visual AI workflows, available as cloud SaaS or self-hosted.
CrewAI is an open-source Python framework and enterprise platform (AMP) for building, deploying, and governing multi-agent AI workflows. It is used by developers and enterprise platform teams.
Ranked near Mindset AI in AI Software
See Mindset AI in the listMindset AI is #81 in AI Software by our score.
