Agent 37 Cloud
Give every customer their own Hermes or OpenClaw agent
931 followers
Give every customer their own Hermes or OpenClaw agent
931 followers
Agent 37 is managed hosting for persistent agents like Hermes, OpenClaw and ClaudeCode. So you don't need to run them on Mac minis or VPS yourself. One API call gives each of your customers their own always-on agent, from $3.44/mo. Founders use it to ship vertical agents to their own clients without babysitting servers.
Love this! But - are you compliant with EU laws?
Under what jurisdiction does your company run and do you have a DPA for EU/GDPR compliance?
Agent 37
Thanks @erik_v We're currently US based but we have a few EU users who are also asking for GDPR compliance. We're working with them to get it ready for it, and will be adding EU servers soon so that it's an option for EU users.
The unit economics here look incredible for SaaS teams wanting to deploy dedicated AI features. At $3.44/mo per customer agent, how do you handle resource provisioning if a client's agent suddenly experiences a massive spike in background workflows or data processing?
Agent 37
@andika_fadhilah Thanks yes. We strive to offer the best unit economics :)
We do have tier limits for now, for example a new account, starting with a tier 1 account can only scale to 10 instances limit. They will need to pre-pay a certain amount before we update it to tier 2 and so on unlocking more instances. This also gives us predictability of scale that we can account for.
Judging by the cost structure, this appears to be serverless. I’m curious about the latency on cold paths and whether usage-based billing will apply later on.
Agent 37
@han_circle The latency right now is 1-2 seconds for bootstrapping cold. For the use cases I'm seeing bootstrap time is not really a P1 concern so not something we're optimizing for.
It's already usage based billing btw, we don't bill the whole $3.74 initially, it's billed on an hourly basis based on how long the instance exists.
The per-customer always-on agent angle is the useful part here. I’ve seen the same “agent infra becomes a babysitting job” problem once tools and permissions differ by client. Curious how you separate tenant secrets/tool scopes when an agent is idle most of the time but still persistent?
Agent 37
@xiaosong001 Each tenant is given their own separate container inside of which all tool calls happen and it has it's own data. So this keeps it separate from each other.
I have been running Hermes in a self-hosted setup with Ollama, and one of the biggest challenges isn't the agent itself, it's maintaining the infrastructure, monitoring uptime, handling updates, and ensuring reliability over time. Agent 37's approach of providing persistent agents through a managed platform is an interesting alternative for builders who want to focus on the product rather than server operations. Curious how much customization and control developers retain compared to a fully self-hosted deployment.
Agent 37
Thanks@jahidshah True. And regarding control, you retain most of the control since you have access to the sandbox and can run commands in it - https://www.agent37.com/docs/agents-api/exec
The interesting constraint here is handoff quality: if each customer gets a persistent agent, the audit trail and failure state need to be as easy to inspect as the API call is to start.
Agent 37
@jimmy_lee12 Agreed, for now we have resource monitor and logs in the dashboard to inspect which serves most of the needs. But we also plan to expose it as an API for more agentic debugging.
Not everyone wants to maintain VPS instances or Mac minis. This looks like a much cleaner path.
Agent 37
@morgan__harriss Very true, thanks!