AI WorkersFor Companies
Role-based AI workers that sit above your ERP and CRM, collaborate across tools, and run real work.
One home for every conversation with every worker — searchable, shareable, always in context.
Automations and reminders on timer rules — workers act on schedules, not just prompts.
Queued
Running
Done
Multiple workers collaborate in parallel — hand off across departments and report back when the job is done.
MiraSales
Qualifying lead
NovaFinance
Matching invoice
ReedOps
Updating inventory
Persistent context across sessions — your company's institutional memory, held by the Companion Brain.
Company memory
How It Works
Give it a role, outcomes and boundaries — like writing a job description, not a prompt.
Link your ERP, CRM, email and files. The worker loads your company's memory and speaks your language.
Workers collaborate, hand off across departments, and report back — over WhatsApp, Email or Telegram.
The Workforce
Every worker is pre-trained for its role. Assign a channel, load your company context, and it starts working.
Workers operate across WhatsApp, Email and Telegram, loading company-specific context from your files and knowledge notes. More expert roles on the roadmap.
The Platform
Assign tasks, monitor health, review performance — a true digital workforce layer. This is software for AI employees.
One home for every conversation with every worker — searchable, shareable, always in context.
Automations and reminders on timer rules — workers act on schedules, not just prompts.
Agent health between check-ins, so you always know your workforce is alive and on-task.
Persistent context across sessions — your company's institutional memory, held by the Brain.
WhatsApp, Telegram, Discord, Gmail — workers meet your team in the channels you already use.
Switch between OpenAI, Claude and Gemini per worker — the right model for each job, no lock-in.
Pricing
Every customer tier maps to a distinct engagement model — from self-serve SMEs to organisation-wide enterprise rollouts.
Early-stage SME
Flat US$20 per agent / month + usage-based credits
Scaling business
Monthly subscription + advisory fee
Medium–large enterprise
Advisory + subscription + premium features, per engagement
Corporate, GLC & large enterprise
Setup + subscription + multi-year contract
When a skilled employee leaves, their knowledge leaves with them. With Hawkin, your business owns its AI workers as proprietary assets — clone one, license it to another company, keep the original, and earn from every clone sale. Workforce knowledge becomes ownable, scalable, and tradeable.
Deploy your first worker this week — and let it prove itself on the work your team shouldn't be doing.
An automation runs a fixed rule. An AI worker holds a role. It has a defined job, company context loaded from your own files and knowledge notes, a channel it works on, memory that persists between conversations, and the ability to hand work to other workers. It behaves less like a script and more like a colleague.
No — the opposite. Hawkin sits above your existing stack as an execution layer. Your ERP stays the system of record, your work tools stay where they are, and the AI workers act across them. Nothing gets ripped out.
Wherever your team already is: WhatsApp, Email, Telegram and Discord, with connectors into your operational systems. You assign a worker a channel and it starts handling work there — no new interface for your staff to learn.
You choose. Hawkin supports multi-LLM switching across OpenAI, Claude and Gemini, so you’re not locked to a single provider or a single price curve.
The first wave covers Sales, Marketing, Customer Service and Admin. Next comes HR, Finance, A/R and A/P specialists, followed by Procurement and Operations. Alongside these, our horizontal systems — CRM, finance claims, warehouse management and product master — are already live.
Starter is self-serve at USD 20 per agent per month plus usage-based token consumption, with an AI credit allowance included. Growth adds guided onboarding and advisory. Pro and Enterprise are quoted per engagement, since they include bespoke builds and org-wide rollout.
A pre-built worker goes live the same day: assign a channel, upload your company context, and it starts handling work. Multi-department orchestration — several workers coordinating across functions — typically takes two to four weeks, most of which is mapping your actual workflows rather than technical setup. Custom workers and bespoke integrations are scoped during discovery.
Every worker is monitored the way you’d monitor staff. The Heartbeat Monitor tracks agent health between check-ins, the Task Engine governs scheduled work and reminders, and dashboard analytics show performance across the org — one pane of control rather than scattered, ungoverned agents.
Yes. A worker you configure is your asset. You can also clone it and license that clone to another business through the Hawkin marketplace — keeping the original and earning from the transaction. Workforce knowledge becomes something you own rather than something that walks out the door.
No. Your files, context and worker memory are yours alone — they’re never used to train models, ours or our providers’. We run on enterprise API tiers where customer data is contractually excluded from model training, and we don’t pool one customer’s data into another’s workers. Full terms are set out in our Data Processing Agreement.
