Finance brokerage processes are known for being clunky and complex. It’s traditionally been very paper-heavy and manually intensive, with opaque and confusing credit policies, and it often takes weeks for borrowers to get an answer on whether their loan is approved.
By using AI to transform the lending process into a seamless, quick, and easy journey, LoanOptions.ai is driving change across the Australian brokerage industry.
The company’s AI-native suite of products, built on MongoDB Atlas, includes five-minute asset finance and home loan application platforms that prefill client details from uploaded documents, as well as an LLM-powered credit policy assistant that provides rapid answers to complex policy questions.
“As a technology company, our goal is to deliver a flawless user experience journey,” said Julian Fayad, Founder and CEO at LoanOptions.ai. “While what we do in the background is very complex at an operational level, for the end-user it has to be easy and seamless.”
Building and innovating at pace and scale
LoanOptions.ai was established in 2020, and in 5 years, scaled to hundreds of thousands of users running millions of loan comparisons: “When we launched, AI wasn’t as mainstream as it is today,” said Fayad. “We were doing something truly disruptive: building a genuine AI-native company, from scratch. For that to succeed, we knew that we needed a database suited for AI and for scale.”
LoanOptions.ai handles complex, unstructured datasets. Each loan application involves layers of sensitive, interdependent data—including IDs, payslips, assets, and liabilities, all in multiple data formats—making each application unique. On the lender side, there are a plethora of complex policy documents to factor in for context.
Processing these required flexibility, integrated vector embeddings, and deep search capabilities that can handle semantic meaning and context, not just text. Traditional, rigid relational database solutions were unable to address these core requirements.
“As a lean, bootstrapped fintech company, we can’t afford to manage a complex, inefficient data architecture. We need to focus on building product features, and we need to move fast,” said Fayad.
LoanOptions.ai identified MongoDB Atlas as the optimal data platform to build upon for the following reasons:
Flexibility: MongoDB Atlas’s document model offered a perfect match for handling complex financial data, such as applications and policies, simplifying LoanOptions.ai’s code and eliminating slow, complex database joins. MongoDB’s flexible schema also enabled LoanOptions.ai to evolve its products and add new features on the fly without the costly, time-consuming schema migrations required by SQL.
Scalability: MongoDB Atlas enabled LoanOptions.ai to confidently handle traffic spikes and scale, from the first user to over hundreds of thousands of users —without ever needing to re-architect the core database.
Built-in security: MongoDB Atlas, hosted on AWS, delivered a native, secure integration with LoanOptions.ai’s AWS Virtual Private Cloud via AWS PrivateLink. This ensures that mission-critical data never travels over the public internet.
Built for AI: MongoDB’s flexible document model is inherently suited to AI workloads. Additionally, MongoDB Atlas provided LoanOptions.ai direct access to a single platform with advanced features like MongoDB Vector Search, enabling the company to build AI-driven tools without managing a separate vector database. This eliminated cost, complexity, and helped speed up LoanOption.ai’s innovation.
Seamless, fast AI-first lending journeys
Powered by MongoDB Atlas, LoanOptions.ai has scaled to process AU$1.3 billion of loan applications annually.
The company’s financial loan platform can pre-fill over 80% of clients’ loan applications, making the whole application process an easy, five-minute experience for clients. LoanOptions.ai has reduced the end-to-end lending and funding process—from the time the client applies to the time the money is in their bank account—from days to 36 minutes, an industry record in Australia.
MongoDB’s flexible schema has simplified how the tech team works: “Our engineers don’t have to worry about all the traditional admin load and maintenance,” said Nirlep Adhikari, Chief Technology Officer at LoanOptions.ai. “With MongoDB Atlas, they are freed up to work on all the valuable AI innovations. That makes a big difference in their job satisfaction, and in our ability to recruit the best talent as well.”
Particularly, using MongoDB Vector Search on Atlas has been a game-changer. In its early years, LoanOptions.ai had to rely on separate, often-disconnected vector databases, which introduced significant operational overhead, data synchronization issues, and added layers of complexity and cost. In 2024, they decided to consolidate by using the powerful vector search capabilities built into MongoDB Atlas. This not only eliminated the need for a separate, siloed database but also streamlined the company’s architecture. MongoDB Vector Search is now what powers LoanOptions.ai’s LLM-powered credit policy assistant, Ask AILO.
MongoDB Vector Search also enables the team to vectorize and recall data from multi-page banking documents seamlessly. This means that Ask AILO can provide instant answers to complex policy questions using data from across 90 lenders.
MongoDB Atlas is also at the core of Synapses, LoanOptions.ai’s proprietary API-first ecosystem. LoanOptions.ai enables other lenders to eliminate the technical debt plaguing traditional finance systems. With over 30 API endpoints, Synapses enables both internal tools and third-party partners to access a sophisticated suite of capabilities. These include intelligent data capture, predictive credit engineering with real-time approval probability, contextual matching, and unified financial lookups.
“By consolidating disparate services into a single API, we help other lenders reduce development costs and time-to-market, so they can maximize their ROI with minimal effort,” said Fayad.
Finally, MongoDB has helped LoanOptions.ai meet stringent security and compliance standards that are indispensable for operating in the fintech industry. This included ISO 27001, an international standard for managing information security.
Building toward even more speed and automation
Looking forward, LoanOptions.ai aims to continue smashing its record end-to-end lending and funding time: “We want to make the whole lending and funding journey as easy as checking out at an e-commerce store,” said Fayad.
The company’s long-term vision includes the development of Supervized Autonomous Loans (SAL)—using MongoDB—to deliver faster, bias-free results in the client’s best interest, while human brokers provide the level of personalization and compliance checks needed for optimal results.
Additionally, by opening up the Synapses API ecosystem to the wider market, LoanOptions.ai plans to progress the entire finance industry: “Imagine if every car owner had to build their own roads to drive on,” said Fayad. “With the help of MongoDB, we build better, faster, more secure roads for everyone to drive on.”
Next Steps
Learn more about MongoDB Atlas.
Learn more about MongoDB Vector Search.