Built Around the Belief That AI Should Serve People
Usaha AI was founded in Kuala Lumpur with the understanding that artificial intelligence is only useful when real teams can act on what it produces.
Back to HomepageA Kuala Lumpur Practice Focused on Practical AI
Usaha AI began as a small consultancy in 2016, working with mid-sized Malaysian companies that were curious about machine learning but had no internal technical teams to guide the process. From the outset, our focus was not on building the most sophisticated models — it was on building models that people in the business could understand, interrogate, and act on.
Over the years, our work has expanded into three core areas: customer intelligence, safe AI experimentation infrastructure, and organisational knowledge systems. Each area emerged from watching the same challenge play out repeatedly — organisations investing in AI and then finding that the outputs don't connect to the decisions people actually need to make.
We are a Malaysian team, working mostly with Malaysian organisations, and that context matters to how we work. We understand the business environment, the workforce dynamics, and the regulatory landscape here in a way that shapes every engagement.
Our Mission
To make artificial intelligence a functional part of how Malaysian organisations work — not a separate system that produces reports no one reads, but something embedded in the daily rhythms of how teams make decisions.
Our Values
- Transparency — We explain what we're building and why, in language your team can follow.
- Restraint — We recommend the simplest solution that solves your problem, not the most advanced one we can build.
- Respect for Data — We treat your data with care, always working within agreed governance boundaries.
- Long-term thinking — We consider how our work will be maintained and evolved by your team after we leave.
People Behind the Work
Each member of our team has worked in Malaysian business contexts and understands the gap between AI theory and organisational reality.
Ahmad Hafiz
Principal AI Consultant
Ahmad leads our customer intelligence practice, having spent a decade helping subscription businesses in Southeast Asia understand their retention dynamics.
Sarah Lim
Infrastructure & Governance Lead
Sarah designs our sandbox environments and governance frameworks, with a background in enterprise security and cloud architecture across Malaysian financial services.
Ravi Nair
Knowledge Systems Specialist
Ravi leads our knowledge management engagements, drawing on experience in taxonomy design and semantic search deployment across professional services firms.
Standards We Hold Ourselves To
Data Handling Protocols
All client data is handled under written agreements that specify what may be used, where it may be stored, and when it must be deleted or returned.
Scope Transparency
Every engagement begins with a written scope that describes deliverables, timelines, and what falls outside the engagement boundaries.
Adoption Support
We stay involved after delivery to support adoption. A system that your team doesn't use is not a successful engagement, regardless of its technical quality.
Independent Review
Our models are reviewed before delivery against quality criteria agreed during scoping. We do not release outputs we are not confident in.
Responsible AI Commitment
We will not build systems that automate high-stakes decisions without human review, and we discuss ethical implications as a standard part of scoping.
PDPA Compliance Awareness
All projects involving personal data are designed with Malaysia's Personal Data Protection Act in mind. We flag compliance considerations early in the scoping process.
AI Integration That Works in Malaysian Business Contexts
Malaysian organisations face a distinctive set of conditions when it comes to artificial intelligence. Data infrastructure varies considerably across sectors. Workforce familiarity with AI tools ranges from deep technical fluency to complete unfamiliarity, sometimes within the same department. Decision-making structures often involve layers of review that AI systems need to accommodate, not circumvent.
Usaha AI has been working within these conditions since 2016. Our AI integration work spans subscription businesses managing customer relationships, professional services firms organising complex institutional knowledge, and technology companies building the internal capacity to experiment with AI safely. In each context, what determines success is not the sophistication of the algorithm — it is whether the organisation can make use of what the algorithm produces.
Our customer retention modelling engagements are built for businesses where churn is a meaningful operational risk. Rather than producing a single churn score, we design systems that surface the specific intervention points your customer success team can act on, calibrated to the data patterns that are actually present in your customer base.
Our AI sandbox environments address a gap that many organisations encounter: the desire to explore AI tools and models without the risk of disrupting production systems or creating ungoverned data practices. We provision, configure, and document these environments in a way that encourages structured experimentation while maintaining the oversight your IT and compliance teams need.
Our knowledge management engagements recognise that the most valuable knowledge in most organisations is not stored in a system — it lives in documents, email threads, and individual expertise. We build AI-assisted systems that make this knowledge searchable, surfaceable, and usable by people who don't know where to look.
Interested in Working Together?
We're happy to discuss your situation informally before any commitment is involved.
Get in Touch