Bared Footwear took the friction out of AI adoption by enabling governed data access
Guardrails for AI on internal systems, without stalling adoption.
Sumesh Suvarna, Bared's CTO, wanted the whole company using AI on the systems they touch every day, and he knew every extra step would cost adoption. But once ChatGPT and Claude are the things reaching those systems, the guardrails you'd build into a UI aren't there. Locking everything down would stall the rollout he was after. Leaving it open was what kept him up.
“We wanted everyone using ChatGPT and Claude to reach our internal systems. But when it's an LLM making the calls, not a UI or API keys you manage, you need guardrails you can trust to sleep at night.”
Sumesh Suvarna
CTO, Bared Footwear
The visibility question sat alongside it. Bared was giving people access to internal platforms with no real way to know how that access was being used. Was AI adoption actually improving across the company, or did it just feel that way? It was a gut call.
Governance that people don't feel.
The gateway gives Sumesh enough visibility to see who's using what, the usage patterns, and which platforms earn their keep. Governance is built in, so some actions can't be taken by everyone and the right requests route to the right person for approval, without the friction that makes people abandon a tool.
“The less the friction, the more the adoption, and that's where MintMCP came in.”
And because Mint runs on its own product, when Bared needs a feature it gets to an outcome fast instead of joining a long ticket queue.
Bared controls what people can do with virtual MCPs and bundles. A large group gets read-only access to BigQuery while a small subset gets write-heavy operations, all through different virtual MCPs pointed at the same platform. For company-wide knowledge, there is a standard bundle every employee gets on arrival, spanning more than 15 connectors from Netsuite and Klaviyo to BigQuery.
“In terms of AI and technology maturity, Mint has upped our game multiple-fold, just by implementing this tool.”
What Bared Footwear told us
Easy to configure, easy to use, easy to see. Sumesh describes the setup as clicking one, two, three to enable a platform across the tools people already use. It's easy for the end user on Claude or ChatGPT, easy for admins to configure, and easy for him to look at the logs and dashboard to see token usage and which tools people lean on most.
“The moment a new employee walks in, they have information across our whole company within a second. I never experienced that before, and Mint helped us achieve this.”
Someone walking in on their first day can pull sales data, marketing data, and transaction information in seconds, all governed by the bundle they're given. Next, Bared is starting to build coworker agents in Slack, its primary place of work, each with its own agent identity tied to the systems it can reach. Sumesh thinks of an agent as a junior engineer: real knowledge, but you still onboard it and train it like a new hire. Early experiments include a technical writer that updates the risk register when someone tags a risk, and an agent that pushes sale-price changes across systems.
“Mint gave us a lot more assurance that what we're doing is secure, and now we know what's happening with agents and MCPs.”
Move from systems opened blindly to AI to every call governed
See how Mint lets you put your whole team on governed AI, with new hires productive on day one and the visibility to back it. Read more stories