BM Agents
The short answer
BM Agents is the runtime we build custom agents on. It provides tool use across your SaaS and internal APIs, role-based access, evaluation suites, guardrails and a full request trace, and it runs hosted or inside your own VPC.
BM Agents wraps your tools, data, and policies in an agent runtime built for enterprise. Tool calling, evals, audit trail, role-based access, out of the box. It is the layer the rest of the suite is built on, and the one we use when a workflow needs something bespoke.
What it does
- Tool use across SaaS & internal APIs
- Role-based access & audit logs
- Eval & guardrails by default
- Hosted or in your VPC
What it connects to
- Internal and third-party APIs
- Model providers, kept swappable
- Vector and search infrastructure
- Identity providers and RBAC
- Observability stacks
Controls
- Evaluation suite gates every release
- Guardrails at input, tool call and output
- Typed tool contracts with defined failure behaviour
- Pinned and revertible prompt, model and tool versions
- Auto-resolve
- 71%
- Median TAT
- 38s
Figures come from delivered engagements and depend on volume, data quality and process maturity. The measurement window behind any of them is available on request.
Seen in production
What to instrument before your first agent goes live
Traces, evals and cost caps are not a phase two. Here is the minimum you need in place on the day you put an agent in front of a customer.
- Issues caught pre-customer
- 83%
- Mean time to diagnose
- < 10 min
- Cost variance
- ±6%
The engagement it sits inside
A product on its own rarely solves the loop. This is the engagement BM Agents is normally delivered as part of.
Production engineeringProduction AI Agent Development
Production AI agent development is the engineering that turns a working demo into a system you can operate: evaluation suites, guardrails, tool integration, observability, rollback and an owner. Branemind builds agents to that standard, and will tell you when a workflow does not need an agent at all.
Questions we get asked
Can BM Agents run in our own cloud?
Yes. The runtime deploys into your VPC, and model provider choice can be constrained where data residency requires it. Decide that before the build rather than retrofitting it.
Are we locked to one model provider?
No. Provider choice is a configuration. Keeping it swappable is one of the reasons the eval suite exists, because it is what lets you compare providers on your own cases.