Cortex24
AI assistant for Bitrix24
Use plain-language requests to work with CRM records and reporting. Explore the examples and discuss your workflow.
FusionETA, based in Kuala Lumpur, engineers AI agents into the systems Malaysian businesses already run — Bitrix24, Xero, WhatsApp and your HR stack — to reduce repetitive work across sales, finance and operations. Start with one process, or connect your teams with a wider rollout.
| Client A — Professional plan | 12 Sep |
| Client B — Standard plan | 19 Sep |
| Client C — Enterprise plan | 27 Sep |
Choose the support your business needs today. Each service starts with the way your team works.
Connect AI to your CRM, accounting and WhatsApp workflows, with approvals built into the process.
Explore AI implementation → 02 / Sell & collaborateOrganise enquiries, follow-ups and business workflows around a shared customer record.
Explore Bitrix24 services → 03 / Manage financesMove your accounting to Xero and connect finance with the rest of your business.
Explore Xero services →Need your systems to work together? Explore integrations and e-invoicing workflows.
Explore AI assistance for CRM, finance and HR, then shape the implementation around your team.
AI assistant for Bitrix24
Use plain-language requests to work with CRM records and reporting. Explore the examples and discuss your workflow.
Invoice and bill automation
Explore the journey from an invoice attachment to a draft Xero bill, with job matching and review built into the process.
HR workflows through conversation
Explore an AI assistant for attendance, claims, leave and payroll questions, built around the way your team works.
For six years we've mapped and automated how Malaysian companies sell, bill, buy and report. AI agents are the next layer on that work — they take on the judgement-heavy, repetitive steps that rules-based automation couldn't handle.
We trace how work actually flows today — inputs, hand-offs, systems, exceptions — and pick the steps where an agent pays back fastest.
Define what the agent reads, what it may change, what needs approval, and which figures must always come straight from your system of record.
We connect the agent to Bitrix24, Xero, WhatsApp or your apps through APIs and MCP servers, then pilot it with a real team on real data.
Your team learns to work with the agent and adjust its instructions. We monitor, tune and extend it to the next process.
Three examples from our client work. Deployment status is shown for each. Explore all ten use cases →
Account managers kept exporting reports to find which renewals were due, overdue or churned.
Outcome: renewal status on demand for the whole team, with every change logged on the deal.
Read this case study →Moving several group companies from a legacy desktop ledger to Xero meant thousands of lines to re-map and reconcile.
Outcome: 57 journals / 669 lines rebuilt and balanced for one entity; bank, AR and AP reconciled to source.
Read this case study →Site teams sent material requests as WhatsApp photos and voice notes; purchasing retyped every one.
Outcome: no retyping between site request and PO; approvers stay in control of spend.
Read this case study →Every agent we ship follows the same engineering rules, whether it answers one team or runs across a group of companies.
Totals, balances and counts are calculated by code from Bitrix24 or Xero. The model interprets the question; it doesn't make up numbers.
Stage moves, invoices, POs and journals are proposed first. A person approves before anything changes.
Each agent gets only the scopes its job needs, using dedicated keys you control and can revoke.
Agent actions are written to the CRM timeline or accounting history, so you can always see who changed what, and why.
SST, MyInvois e-invoicing, withholding tax, EPF and SOCSO rules are designed into the process, not left to the model.
Agents take the repetitive steps; your team keeps the decisions. We train staff to supervise and adjust them.
One high-volume process, one agent, inside the tools you already use.
A set of agents across sales, finance or operations, sharing one integration layer.
Agents across multiple entities and systems, with governance to match.
We deploy and run your new AI agent for you — no separate bills for AI usage, hosting or support.
Final quote depends on the process and integrations involved. Ask us to confirm usage allowances, support scope, taxes and any third-party subscriptions in your proposal.
Get a quoteThat's why our agents sit on six years of certified Bitrix24 and Xero implementation work — clean pipelines, clean ledgers, clean data.
A chatbot answers questions. An AI agent also takes action inside your systems — reading CRM records, drafting purchase orders in Xero, moving deals, posting notes — within the permissions and approval rules you set.
Most of our agents are built on Anthropic's Claude, chosen for its reasoning and tool-use reliability. Where a platform has its own model, such as the Bitrix24 AI router, we use it where it fits.
Yes. SMEs usually start with one agent on one high-volume process — renewals, purchase requests, document capture — running inside tools they already pay for such as Bitrix24, Xero and WhatsApp.
Deploying a new AI agent with FusionETA starts from RM3,000 per year, inclusive of API token usage, a 24x7 server, training and maintenance. The final quote depends on the process and integrations involved — see pricing.
Figures come from your system of record, not the model. Write actions require confirmation, every change is logged, and agents only get the permissions their job needs.
A focused single-process agent typically moves from process review to a working pilot in a few weeks, depending on integrations and data readiness.
Book a free AI process review — we'll map the workflow and outline the agent, guardrails and integrations it needs.