24/7 support
Immediate answers at any hour, with no extra headcount and no night queue.
Atena
Support that resolves on first contact.
Atena serves the player in the logged-in WebChat, answers from the knowledge base approved by the operator and escalates to a human when the case calls for judgement.
WebChat · logged-in siteAtena · AI agent
demo · sample dataMy withdrawal has not arrived yet.
R$ 500 withdrawal approved at 14:32. The bank releases within 2 hours.
Perfect, thank you!
12sto first reply
0in the human queue
1touch to resolve
Resolved without a support repand logged to the trail instantlyresolved
A queue that will not move, answers off script and evidence scattered around pull the team away from what matters.
Atena resolves with sources.
Cases that need judgement reach people ready to handle.
Sourced answers
Every answer comes from the knowledge base approved by the operator.
The rulebook is yours
Responsible gaming and self-exclusion follow the policy you define under Law 14.790/2023. Cortex applies it in every conversation and records that it did.
Structured handoff
A sensitive case goes up to a human with the whole conversation context along.
Presence that never goes offline.
The player is served at any hour in the logged-in WebChat, with the same tone and the same criteria — with no dependence on who is on shift.
Support coveragechats per hour · last 24 hours
demo · sample data00:0006:0012:0018:0023:00
Tuesday, 3 a.m.same tone and criteria as business hours
12s
Sunday night peakno waiting queue, no extra shift
14s
Answer being preparednot shown to the player yet
Your deposit limit is R$ 300.00 per week. To lower it, I am passing you to someone on the team now.
Mandatory checks
Operator-approved basesource cited in the answer log
Contract limit and capnothing promised outside the rule
Responsible gamingNêmesis checked before the reply
Cleared to sendconfidence 0.94 · off rule, she stays silentby the rules
Answers that follow the rules.
Atena answers from the operator's approved base and checks Deméter and Nêmesis before speaking. No improvising on withdrawals, limits or bonuses.
Evidence ready for audit.
Every decision goes to a WORM log with a chained hash and redacted PII, retained for five years. When the regulator asks, the proof is already assembled.
WORM trailevery decision becomes a record the moment it happens
recordingEventHash
Answer sent to playeratena.message_sent · atena
a71c4e02…
Handoff triggered by Atenaatena.session_handoff_triggered · atena
4e8d3b57…
File opened by the support repatendimento.ficha.accessed · marcos.r
9c05f7ae…
Write-only · chained hash · retained 5 yearsa71c4e0293f8b6d15ac47e0b3fd825619ce74a0…
Handoff to Backoffice#064B17A2 · Tatiana Macedo Ferreira
escalated
Escalation reason
Exception request beyond AI authorityAtena does not decide contract exceptions
What goes along
The whole conversation12 messages — the player repeats nothing
Profile and behavioursensitive data masked by default
Decision trailwhat Atena tried, and why she stopped
Marcos R. took over in 40swith the context already on screenwith a human
The human team on what is human.
There are six triggers that take the conversation to a human: responsible gaming, emotional crisis, regulatory threat, resolution failure, an explicit player request and an unresolved limit in the conversation. The case reaches the operator with the whole context, with no repeated questions.
Connect your operation
We plug Atena into the WebChat and the backoffice you already run today.
Tune the agent
Your house rules become behaviour: tone, limits, authority and knowledge base.
Go live with hypercare
Go-live monitored for 15 days, with resolution and escalation metrics in plain sight.
Frequently asked questions
How long does it take to put Atena live?
The rollout track takes 15 to 30 business days, counted from the delivery of your access and integrations. Go-live includes 15 days of monitored hypercare.
01Connect access and integrations.
02Set the rulebook and approved knowledge base.
03Go live with support.
Does Atena replace the support team?
No. Atena handles repetitive volume within the approved rulebook and takes cases that require judgement to a person. The team receives the conversation context and evidence already gathered: emotional crisis, regulatory risk, a contract exception or a player's request.
HandlesRepeated questions, within the approved rulebook.
EscalatesCases that require human judgement.
What does the record for the regulator look like?
Every decision goes to a write-only log: the database will not allow altering or deleting a row already written, each record carries the hash of the previous one, and tampering with one of them breaks the whole chain — which makes alteration detectable, not impossible. PII goes in redacted, retention is five years and, at the end of it, the data is disposed of as the LGPD requires. The trail is exportable for audit.
Message received and rule consulted.
Response recorded with decision context.
Hash chained and trail available for export.
Do I need to replace my current WebChat?
No. Atena runs on top of the WebChat and backoffice you already operate. Integration is by API and events, with no stack change.
Current WebChatPlayer conversation.
AtenaApplies the approved rulebook.
Current BackofficeReceives context and evidence.
How does Atena handle responsible gaming and self-exclusion?
Atena applies the Nêmesis policy: it detects risk signals in the conversation, routes limits, pauses and self-exclusion, and surfaces your jurisdiction's official support channels in a crisis. Basis: the gaming regulation applicable to your operation.
Signal in the conversationAtena identifies the context.
Policy appliedLimit, pause or self-exclusion.
Human engagedWhen the case needs human care.
How much does it cost and how is it billed?
Per credit — 1 credit equals 1 Atena answer. Each plan includes a monthly fee, a setup fee equal to the monthly fee and a monthly credit pool. Usage above the pool becomes a top-up.
1 creditEquals one Atena response.
Monthly poolVolume contracted for the cycle.
Top-upCovers usage above the pool.
What if the credit pool runs out mid-month?
All agents pause, compliance included, until top-up. No conversation is cut mid-way: whatever started is finished, and your human team keeps working from the Backoffice because human service consumes no credit. Usage is measured by cycle and agent and remains available for review at any time. The operator receives automatic alerts when usage reaches 80%, 95% and 100% of the contracted pool. The legal duty of responsible gaming and anti-money-laundering remains the operator's, with or without credit.
Open conversationEnds without being cut off.
AgentsPause until top-up.
Human teamKeeps working in Backoffice without consuming credit.