Every question. One page.
The general questions, the product details, the pricing math, and the vertical-specific ones from logistics, finance, and customer ops — gathered in one scroll so you can find what you need without opening six tabs.
The five we always get.
No marketing answers. If yours isn’t here, the demo will get to it in ten minutes.
01How long to first agent live?
Three weeks, end to end. Week 1 we shadow your operators and pick the first agent. Week 2 we co-build it. Week 3 it runs in your inbox, watched closely, with a manual override for everything.
02What models do you run on?
Frontier (Claude · Sonnet / Haiku) for reasoning, your private retrieval for context. We pick the smallest model that hits the bar for each step — that's how we keep the per-run cost honest.
03Does the data leave my tenant?
Only if you say so. Dedicated cloud + private VPC on Enterprise; for everyone else, EU/US tenants you choose. We don't train on your data. Audit log shows every byte the agent touched.
04What if the agent gets it wrong?
Then your operator gets it. Every step is logged, the receipt explains what the agent saw and what rule fired, and the human picks up the override. You see the failure rate per agent on the dashboard.
05Do we have to rip out our TMS / CRM / helpdesk?
No. Lumin posts into the systems you already run — that's the whole point. We connect via your APIs and your auth; the agents become a layer on top.
Product, more deeply.
The questions an engineering leader actually asks before signing.
01What models can I use?
Any frontier model (Claude, GPT, Gemini) and most open-source ones via your inference provider. Pick per step, swap any time. Lumin never locks you in.
02Can I edit an agent that’s running?
Yes — branch, edit in the Builder, run eval, merge. Production stays on the previous version until you promote.
03How do I bring my own tools?
REST endpoints, OpenAPI specs, and webhooks are first-class — Lumin generates a tool node from any spec. For deeper hooks, write a custom node in TypeScript.
04What about latency?
Median agent run is 1.1s. p95 around 3.4s. Each step has its own budget; the runtime chooses caching + fallbacks automatically.
05How does this differ from a workflow tool?
Workflows assume the path. Agents assume the goal. Lumin gives you both: deterministic nodes where you want certainty, model-driven nodes where you want judgment.
06Do you support on-prem?
On Enterprise — dedicated cloud + private VPC, with optional BYO KMS. Pure-on-prem is not officially supported; talk to us.
The six on the invoice.
01What counts as a run?
One end-to-end execution of an agent: input → parse → plan → tool calls → output → receipt. A single ticket = one run regardless of how many model calls happen inside.
02What if I go over my monthly cap?
Overage at the same per-run rate as your tier. We email you at 80% and 100%, and you can hard-cap if you want. Never a surprise bill.
03Can I pay in USD / GBP / annually?
Yes to all three. Annual saves 12%. Stripe + invoice both supported.
04Can I bring my own AI provider key?
Growth + Enterprise: yes. We pass through model costs at our negotiated rates by default; bring your own for full custody.
05Is there a free trial?
14 days, no card. Includes Starter limits + a sandbox dataset to try the Builder on.
06Refunds?
Yes — pro-rated, no questions, within 14 days of any month's invoice.
What logistics leaders ask.
01Does it work with my TMS?
McLeod, MercuryGate, Aljex, AscendTMS, Tai, Turvo, Trimble, FreightPath, NetSuite, SAP — all native. Anything else via REST + webhook.
02Can it pick the carrier or just suggest one?
Both, your call. Default is suggest + auto-book if margin ≥ X%. Above the cap, it asks the dispatcher. Every decision is in the receipt.
03What about EDI?
Native X12 (204, 990, 214, 210) and EDIFACT. Drop-ins via SFTP or AS2. We map to your trading partner specs in onboarding.
04Phones, really?
Yes — voice agent for carriers calling in. Same brain, same receipt, transcripts attached.
What finance leaders ask.
01Is it SOC 2 + GDPR?
SOC 2 Type II in production. GDPR + DPA signable. EU and US tenants. Dedicated cloud on Enterprise.
02Can it post to NetSuite + SAP?
Yes, both native. QuickBooks, Xero, Sage, Oracle Financials, Workday — all native. Custom ERPs via REST.
03What happens to flagged invoices?
They go into a human queue with the rule that fired, the variance, and the receipt link. Your team approves or escalates.
04Can the model see PII?
Yes, but it never leaves your tenant. We don't train on customer data, ever. Retention is per-tier; full purge on request.
What customer-ops leaders ask.
01Will it sound like us?
Yes — Lumin learns your tone from your existing reply corpus and your style guide. You approve every prompt before it goes live.
02What about complex cases?
Anything outside policy, anything emotional, anything escalated — hands off to a human with the full thread + the receipt. Never invents an answer.
03Does it integrate with Zendesk / Intercom / Front?
Yes, all three native. Salesforce Service Cloud, Kustomer, Gorgias, Freshdesk too. Your tickets stay where they are.
04How fast is the chat agent?
First token in 200-400 ms. Full reply under 1s for 92% of conversations. Tracked + visible in Reports.
Ten minutes. Your tickets.
Anything not covered above — we’ll answer it on a working session against a sample of your inbox. No deck, no pitch, no card.