Autonomous agents with memory and context — patent filed with the USPTO

PT EN
Agent on WhatsApp Become a Reseller Client Area
Agentic platform

Agents that decide, orchestrate, and operate

There is no decision tree to draw. The agent is given a personality, knowledge, and tools — your ERP's APIs, your company's knowledge base, your official channels — and it chooses what to do on every turn. Including concluding that nothing needs to be done.

Talk to a specialist Explore the platform

Memory and context for autonomous agents: patent filed with the USPTO in 2025. Meta Business Partner, AWS Partner, and NVIDIA Inception.

An executive working on a laptop in a corporate office at night, with the Hal-AI brand projected in blue light on the wall behind her.

Credentials and partnerships

USPTO patent

Filed in 2025

IEEE Senior Member

Marcos Alves, founder and CEO · May 2025

NVIDIA Inception

Since 2024

Meta Business Partner

Official WhatsApp Cloud API

AWS Partner

Cloud infrastructure
Meta Business Partner AWS Partner Network NVIDIA Inception Program

Programs Hal-AI takes part in. Trademarks belong to their respective owners.

Where to start

Two ways into the same technology

An airline's operation and the WhatsApp line of a neighborhood store have very different demands. The engine is the same; the front door is not.

For your enterprise

Enterprise agentic platform

Agents that handle service at scale and orchestrators that run business routines across your systems, with a full audit trail of every run.

  • Squads that wake up on schedule, query your ERP, and decide what to do
  • Official WhatsApp, webchat, SIP voice, and email — all with the same brain
  • CRM, call center with human takeover, campaigns, and operational auditing
  • Integration with your REST APIs, MCP servers, and a versioned public API
Talk to a specialist
For your business

A virtual employee on your WhatsApp

The agent serves your customers and reports back to you. Pre-trained on your company website, available 24/7, by text, audio, and image.

  • Retail: a virtual sales rep that checks the catalog and follows up
  • Healthcare: a virtual receptionist that books, confirms, and reminds patients
  • Real estate: a digital agent that answers every lead in seconds
  • Education: an enrollment assistant that qualifies and books the trial class
Create your free agent
A whole division

Hospital patient service and management

A hospital is not a large clinic. A patient writes at eleven at night to ask whether they can eat before the exam, the weekly grid has thousands of slots to hand out, and the money for that visit only lands after it clears the payer, the batch, and the denial. There is an entire section of the product for this.

Patient service and booking on official WhatsApp

The patient asks on the hospital's official number and gets an answer: exam prep, which plans are accepted, what documents to bring, the address, the opening hours. When it is time to book, the agent checks real availability, offers the open slots, and writes the appointment into the hospital's official schedule — through the API of the system it already runs, never into a parallel calendar. Rescheduling, cancellation, and confirmation take the same path.

Clinical questions, delicate cases, or an explicit request to talk to a person go to the team with the whole conversation attached. The agent gives no diagnosis and no clinical guidance — and says so to the patient.

See patient service · appointment booking

Schedule and capacity

The full funnel of the day, from the slot offered to the patient seen. Where the grid goes unused, where the queue grows by specialty, and which bookings carry the highest no-show risk.

See schedule and capacity

Revenue cycle

Billing, denials, claims transmission under the Brazilian TISS standard, and coding compliance between what was delivered and what was charged. The discrepancy shows up before the batch leaves, not after the remittance comes back.

Billing and denials · transmission

Dashboards that arrive on their own

A Squad wakes up at the scheduled time, reads the hospital's systems through the APIs it was given, writes the daily briefing with the arithmetic shown behind every figure, and emails it to a closed list of executives. Nobody has to request the report, and nobody has to open a dashboard to know how the day went. For a one-off question, the manager asks the HAL Copilot in plain language and gets the table on screen.

See dashboards and indicators · the whole section

Hal-AI is a Meta Business Partner: the conversation runs on the hospital's own official number, through the WhatsApp Cloud API. Every booking is written into the official schedule through the API — the agent keeps no calendar of its own, and every run is logged.

Explore Hal-AI for hospitals How booking works
The technical difference

What changes when the agent decides

Flow-based automation works as long as the customer follows the script. An agent with memory and tools does not depend on that.

Flow-based automation

  • A hand-drawn decision tree: anything unforeseen breaks it
  • Every new channel means building the bot all over again
  • The conversation starts from zero at every contact
  • Fire and forget: nobody checks whether that customer already replied
  • Reporting is whatever the dashboard can export
  • When a human steps in, you end up with two parallel conversations

Hal-AI agent

  • Receives a goal, knowledge, and tools, then decides on every turn — including not to act
  • The same prompt, the same tools, and the same memory on WhatsApp, on your site, and on voice
  • Vectorized history, episodic facts, and customer traits: the conversation picks up weeks later
  • Agent calls agent (A2A) and gets the answer back: “the customer already confirmed yesterday, so I didn't send it”
  • The agent writes the analytical document, generates the PDF, and sends it to an approved list
  • Taking over pauses the bot; closing the ticket hands the conversation back with the full history intact

This is not a chatbot running a flow. It is an operational agent trained to work.

An AI humanoid reaching out to touch a Hal-AI chip mounted on a circuit board USPTO2025
USPTO · 2025

The memory that keeps the thread

Patent Autonomous Agent Memory Framework Inspired by Human Cognition

Hal-AI's core technical asset is a memory and context framework for autonomous agents, inspired by human cognition. In plain terms, it addresses three things: the agent's short-term and long-term memory, the retrieval of the context that actually matters at that moment, and task continuity from one conversation to the next.

That is what makes it exclusive. It is the difference between an agent that holds the thread of the conversation — remembering who the customer is, what was already agreed, and what is still open — and an automation that starts over at every message.

In practice: history is vectorized through embeddings, a background worker extracts stable facts and customer traits, and the agent carries a sense of time. The same technology powers text-based service, webchat, real-time voice, and the orchestrator agents — which is why switching channels does not mean reconfiguring the bot.

The framework is already applied by companies in logistics, aviation, healthcare, and banking.

Understand the agentic platform

The platform

Six pieces that work together

From the agent talking to your customer to the orchestrator running the business routine at six in the morning.

Squads — orchestrator agents

A master agent with no channel of its own. It wakes up on schedule, queries your ERP through its API, decides what needs to happen, and talks in natural language to the channel agents so they handle each customer one by one. It publishes analytical documents with tables and charts and emails them to a closed list of recipients. Every run has an auditable history, a simulation mode, and a stop button.

See Squads

Memory and context

The agent does not start from scratch with every conversation. Vectorized history through embeddings, episodic memory of stable facts, customer traits, and a sense of time.

It is the same technology on WhatsApp, on webchat, on voice, and in the orchestrator — and it is the subject of the patent filed with the USPTO. The memory screen shows what is stored per channel and lets you clear it.

How the memory works

HAL Copilot

The in-house assistant for whoever runs the business: it queries your own company data in read-only mode, answers with a table and a chart on screen, and exports to PDF. Managers assign the task, close the page, and get a notification when the result is ready.

See the Copilot

Channel agents

A name, a personality, a voice, and real tools: send media and interactive options, request a location, create a WhatsApp template, search the Knowledge Vault, read a website, schedule a task, hand off to a human.

Configure an agent

Security

Controls designed for autonomous agents: curated, parameterized queries instead of free-form SQL, an allowlist of what the agent may do, a closed recipient list checked twice before anything is sent, and refusal of an incomplete document — never a silent trim.

See the controls

Integrations and API

The REST APIs you already have become tools for the agent: register the endpoint, the authentication headers, and a description, and it starts querying and acting in that system mid-conversation — ERP, medical records, order management. There is per-agent support for MCP servers, an importer that reads the source documentation and lists the endpoints, and a versioned public API — with scopes, pagination, and idempotency keys — that even exposes triggering and monitoring Squad runs.

See integrations and API

The six pieces do not live on separate screens: whoever administers the account opens the console and sees everything on a single dashboard — how many channels are up, how many agents are answering, how many conversations are open, and which Squads wake up today.

Hal-AI · Operations dashboard Home
Operations dashboard Today at a glance: channels up, agents answering, conversations open, and the routines still to wake up.
4 channels connected
All activeMeta Cloud API
Two official WhatsApp numbers, the website webchat, and one voice extension over SIP.
7 agents active
Answering1 paused
Customer service, scheduling, and post-sale — all with long-term memory switched on.
38 open conversations
In progress5 with a person
Five were taken over by a person; in those chats the agent is paused.
3 Squad runs today
1 runningnext 2:00 PM
ran today 6:00 AM next today 2:00 PM 2 in simulation schedule on
This week's volume — conversations per day source: operations dashboard
Exam Confirmation Squad running · on schedule Vitoria Clinic · 18 cases read · simulation mode
Campaign guardian slowed the sending pace Heads-up Aurora Store · number quality under watch
HAL Copilot work finished Ready Meridian Freight · Route 12 analysis available as PDF

The dashboard is the first screen after login. The four cards at the top answer at a glance whether the operation is up: channels connected, agents active, conversations open, and Squad runs due for the day. The chart compares the volume of the last seven days, with the weekend dimmed so it does not distort the reading. Below it, the alert strip gathers what needs attention right now — the routine that is running, the campaign the guardian slowed down, and the work the Copilot has just delivered.

Both sides

One agent serves the customer and reports to the people running the business

On one side, the people who buy, book, ask, and complain. On the other, the people who need to know what happened without opening a single dashboard.

“Customers” badge “Team” badge
  • For the customer: an answer in seconds, 24/7, by text, audio, and image, on official WhatsApp.
  • For the operator: a summary of the day's activity, bottlenecks, and open items, plus operational reminders the agent schedules for itself.
  • For the team: when someone takes over a conversation, the bot pauses in that chat; once it is closed, the agent resumes.

This is not a chatbot running a flow. It is an operational agent trained to work.

Customer service online
Hi! My flight is delayed and I'm going to miss my connection. Can it be rebooked?9:12 AM
I checked here: with the delay, the 6:00 PM connection is no longer possible. I can rebook you on the 9:40 PM flight to the same destination. Shall I confirm?9:12 AM
Yes, please confirm.9:13 AM
Done. I've sent your new boarding pass in this chat. I can also continue in Spanish or Portuguese if you prefer.9:13 AM

Connecting to inventory, calendars and back-office systems happens through the integration of your company's APIs.

Channels

One brain, four front doors

Messaging, website chat, and real-time voice all run on the same prompt, the same tools, and the same memory.

Official WhatsApp

Official numbers through the Meta Cloud API, with assisted onboarding, WABA activation, and templates created and approved inside the platform.

Webchat on your site

Embed, shareable link, and QR code. Attachments, photos, recorded audio, and handoff to a human — and the conversation lands in the same CRM.

SIP voice

The customer calls and the AI answers in real time: it talks, queries the same systems, transfers the call, and hangs up. Every call is transcribed.

Email and push

Reports and documents published by the agents arrive by email; a browser notification tells you when the work you asked for is ready.

The four front doors all lead to one place. In the inbox, WhatsApp threads, webchat conversations, and calls answered by the AI show up in the same queue — and every row says who is handling it at that moment: the agent or a person on your team.

Hal-AI · Inbox All channels
Inbox 38 conversations open right now across all three channels, with the owner of each one in plain sight.
Queue · Support
Single conversation queue: channel of origin, who is answering right now, last message, and wait time
ConversationChannelWho is answeringLast messageWaiting
Aurora Store WhatsApp Person · Marina “order 4471 arrived with the wrong item” 6 min
Vitória Clinic Webchat Agent · Scheduling “can I move Thursday's follow-up?” 0 min
Meridian Freight Voice On a call Route 12 · transcript being recorded 2 min
Horizon School WhatsApp Agent · Enrollment “is there a spot in Saturday's trial class?” 0 min
Aurora Store Webchat In the queue “do you deliver same day?” 1 min
AI paused automatically in this chat — taken over by Marina (Team Lead)
Same memory across all three channels the conversation picks up where it left off Memory on Aurora Store · webchat on Tuesday · WhatsApp today · same history

What you are looking at

Up top, the whole queue and the pills counting where the open conversations came from — WhatsApp, webchat, and voice. There is only one table: each row shows the customer's channel, who is handling the conversation at that moment, and how long they have been waiting. The amber banner spells out in words what the first row already flags: that chat was taken over by a person, and the AI is paused in that one thread only. The call in progress sits in the same list, transcribed turn by turn.

Why it matters

Nobody has to open three dashboards to see how big the queue is, or find out too late that the agent and a person both replied in the same chat: assigning the conversation is what pauses the bot, and the badge naming who is answering stays written on the row. Closing it hands the conversation back to the AI with the full history — including whatever the person agreed to along the way.

Only at Hal-AI

Not three lookalike bots: the same brain

The agent that answers on WhatsApp is the same one that answers in your website chat and the same one that picks up the SIP call — same prompt, same tools, and the same long-term memory covered by our patent filing with the USPTO. That is why a customer who wrote in the webchat on Tuesday and came back on WhatsApp today continues the conversation instead of starting over, and why adding a channel is not rebuilding the bot all over again.

Published cases

Where it is already running

Aviation, pharmacy retail, and payroll-deducted lending — three operations with very different demands on the same platform.

+40 languages supported on the airline's WhatsApp service
8000 pharmacies within the integrated chain's potential reach
3 cases published in 2025: aviation, a pharmacy chain, and payroll-deductible credit
A commercial aircraft on the ground, seen from the terminal window.

Aviation

A major Brazilian airline serves passengers over WhatsApp: check-in, disruption handling, and multilingual support in more than 40 languages.

See the case
A pharmacy aisle lined with shelves of medication.

Pharmacy

Hal-AI Agentic integrated with Alpha 7 at one of Brazil's largest chains, with a potential reach of 8,000 pharmacies.

See the case
A professional reviewing financial charts on a screen.

Financial services

A partnership with PHNIX Card: payroll-deducted lending with end-to-end journey traceability over the WhatsApp Business API.

See the case
See all industries AI consulting

Sectors served and credentials

Aviation Financial services Retail Logistics Healthcare HR Education Real estate USPTO patent 2025 IEEE Senior Member NVIDIA Inception Meta Business Partner AWS Partner 2CW group

Choose your path

If the conversation is about operations, systems, and scale, talk to a specialist. If it's about putting an agent to work on your business's WhatsApp today, start with the small-business track.

Talk to a specialist Create your free agent