AI Coding Is Becoming a Team Sport, Not a Single-Tool Bet
If you run a Malaysian SME, you may not have a full-time software team, yet your business still depends on technology. You need a reliable website, customer records, payment processes, stock visibility, staff scheduling, reporting, and integrations between the systems you already use. When one process breaks, you may end up waiting for an external developer or asking an already busy employee to find a workaround.
That is why the growing competition among AI coding companies matters to you. The issue is not whether your company should build the next major AI product. It is whether better and more varied AI development tools will make it easier to improve your own operations without being trapped by one vendor or one technical approach.
Cognition, the company behind the AI coding assistant Devin, announced a US$2 billion funding round at a US$48 billion valuation. Its annualised run-rate revenue reportedly grew from US$492 million to US$900 million after its previous fundraising. Source: TechCrunch
TL;DR
AI coding is attracting several serious players, so you should expect more choice rather than one tool controlling the market.
For your business, the practical opportunity is faster, more flexible automation—but you still need clear processes, human review, data controls, and a sensible way to test tools.
What This Means
AI coding assistants help people create, review, explain, and modify software using natural-language instructions. Instead of writing every line of code manually, a user can describe a task such as, “Create a dashboard showing unpaid invoices by customer and ageing,” and the tool may produce an initial working version.
These systems are becoming more capable because they can handle larger development tasks. Some can inspect an existing codebase, suggest changes, run tests, identify errors, and complete parts of a project. They do not remove the need for experienced people. Software still needs testing, security checks, proper access controls, and someone accountable for the final result.
The Cognition announcement is significant because investors appear to believe that AI coding will support multiple strong companies. Cursor, another coding assistant, had reportedly exceeded US$2 billion in annualised revenue during earlier funding discussions and later agreed to sell to SpaceX for US$60 billion. Source: TechCrunch
In simple terms, different tools may be better at different jobs. One may be strong for professional developers, another for building internal applications, and another for connecting business systems. That creates room for SMEs to select tools based on their actual workflow instead of accepting whatever platform is most popular.
The useful question is not “Which AI tool is the biggest?” It is “Which business process can you improve safely, measurably, and repeatedly?”
How This Applies to Malaysian SMEs
For a Malaysian wholesaler or distributor, AI-assisted development could help connect order forms, inventory records, delivery updates, and customer notifications. You might use it to create a simple internal screen that flags low stock, identifies delayed orders, or groups customers by purchasing activity. The value is not the code itself. The value is fewer manual checks and a clearer view of what needs attention today.
For a restaurant group, clinic, tuition centre, or service business, the same approach could support booking and follow-up workflows. A small internal application might collect enquiries from several channels, assign them to staff, and remind your team when a customer has not received a response. You could also ask an AI coding tool to document how an existing workflow operates, making it easier when a staff member leaves or when you change systems.
For professional firms such as accountants, agencies, contractors, and consultants, reporting is often a hidden drain on staff time. AI-assisted development may help turn information from spreadsheets, forms, and project tools into management dashboards. You could track outstanding tasks, service turnaround, renewal dates, or document status without repeatedly combining files by hand.
Malaysian SMEs also operate across different communication and administrative needs. You may have customers using WhatsApp, staff using spreadsheets, and management relying on accounting or enterprise software. A development assistant can help a capable automation partner create connectors between these systems. However, you should not assume that an AI-generated connection is automatically safe or compatible with your business requirements.
Data handling deserves particular attention. Customer identification details, employee information, transaction records, and supplier documents should not be pasted casually into public AI tools. You need to know where information is processed, who can access it, how long it is retained, and whether your provider can use it for training. A small business can still suffer serious operational and reputational harm from weak controls.
What the Numbers Tell You
| Reported item | Figure | Why it matters to you |
|---|---|---|
| Cognition’s new funding | US$2 billion | Shows strong investor confidence in AI coding development |
| Cognition’s reported valuation | US$48 billion | Indicates the market expects substantial future demand |
| Cognition’s annualised run-rate revenue | US$492 million to US$900 million | Shows rapid commercial adoption, although the calculation method may vary |
| Cursor’s reported annualised revenue during funding discussions | More than US$2 billion | Suggests more than one AI coding provider can attract large-scale usage |
| Cognition’s reported enterprise customers | Mercedes-Benz, NASA, Goldman Sachs, and Citi | Shows these tools are being evaluated beyond small experiments |
All figures and company details in this table are reported by TechCrunch in the source article. Source: TechCrunch
Practical Takeaways for Your Business
- Start with one repetitive process. Choose a task such as invoice follow-up, stock reporting, lead assignment, or appointment reminders. Avoid beginning with a large system replacement.
- Write down the current workflow. List who does each step, what information they use, where delays occur, and what result you expect. Clear instructions improve automation projects.
- Ask for a working prototype. Before committing to a broad rollout, test a small internal dashboard, form, notification flow, or approval process.
- Keep a human approval step. AI-generated code and automated decisions should be reviewed, especially when they affect customer records, payments, compliance, or employee information.
- Check data permissions. Confirm whether the tool stores your prompts, code, uploaded files, and customer information. Establish rules for what staff may submit.
- Require documentation. Every useful automation should include instructions, ownership details, access information, and a plan for future changes.
- Compare tools by business outcome. Measure response time, error rates, staff effort, completion rates, or visibility—not how impressive the software demonstration looks.
- Avoid depending on one individual. Make sure at least one business owner or manager understands the workflow, even if an external specialist builds it.
Questions to Ask an Automation Provider
Before approving an AI-assisted development project, ask what happens to your data, where the application will run, how access will be managed, and who will fix defects. Ask whether the proposed solution can export your information in a usable format if you change providers.
You should also ask how the provider tests the application. A responsible process should include sample records, failure scenarios, permission checks, and a clear sign-off stage. If the answer is only “the AI has generated the code,” that is not enough. You are responsible for the business process, so you need evidence that the process works.
Finally, ask what happens when your operations change. Your tax workflow, approval structure, product catalogue, or customer communication method may change over time. A useful solution should be maintainable rather than a one-off experiment that only its original creator understands.
The Bigger Picture
The competition between AI coding companies suggests that software creation may become more accessible to businesses that previously could not justify a large development project. More providers may lead to better features, broader integrations, and tools designed for different types of users.
That does not mean every SME should build custom software. Sometimes a well-configured existing platform is the better choice. Custom development makes more sense when your process is distinctive, your current tools do not connect properly, or manual work is creating recurring operational problems.
The long-term advantage will come from knowing your processes well. Businesses with clean information, documented workflows, and clear ownership will be better prepared to use new development tools. Businesses with scattered data and unclear approvals may simply automate confusion faster.
For you, the sensible approach is to watch the market without chasing every announcement. Select one operational bottleneck, define the expected result, protect your data, and test a small improvement. If it works, expand gradually. If it does not, you have learned something without putting your entire operation at risk.
AI coding is likely to give Malaysian SMEs more ways to improve internal systems. The winners at your level will not necessarily be the companies using the most advanced tool. They will be the owners who choose a practical problem, set sensible controls, and turn a working solution into a repeatable business habit.
Ready to Streamline Your Operations?
Your business should run itself. AutoRunBiz deploys AI agents to automate your daily operations — WhatsApp orders, invoicing, customer follow-ups, and accounting. Book a free 15-min ops audit to see where automation fits your business →