In an exclusive interaction, Sharad Agarwal, Chief Operating and Technology Officer, U GRO Capital, discusses the company’s recent digital initiatives, including AI/ML-driven tools for credit assessment, API integrations, and digital onboarding to streamline lending processes and enhance customer experience. He emphasises the role of GenAI in content creation, product development, and customer service, while also addressing challenges in implementing AI/ML solutions, such as data quality, scalability, and regulatory compliance. Additionally, he highlights cybersecurity measures and plans to use AI for credit underwriting, reflecting UGRO’s commitment to tech-driven efficiency and outreach to remote MSMEs.
Here are the edited excerpts:
Please take us through some of the recent digital initiatives that you have implemented. What has been the impact?
As a DataTech company, we are deeply committed to leveraging data and technology to streamline our lending processes and enhance customer experiences. From origination to credit, analytics, operations, and collections, data and technology form the core of our operations. With over 25 API integrations, automated policy approvals, machine learning capabilities, a Business Rule Engine (BRE), and customised sourcing modules, we provide a seamless lending journey. These tools enable us to offer in-principle loan approvals within just 60 minutes.
Recent digital initiatives have further strengthened our offerings. For instance, our Retailer Financing app provides MSMEs with tailored working capital solutions, while our specialised Loan Origination System (Skaleup) is designed to serve micro-segments. Additionally, our mobile app GRO X simplifies the MSME onboarding process with features like instant credit limit disbursement. We’ve also implemented video-based due diligence and KYC processes, along with face authentication for secure e-signing. Our AI/ML-driven GRO Score 3.0 model assesses creditworthiness using banking, credit bureau, and GST data.
These initiatives have reduced turnaround times, improved operational efficiency, and enhanced the customer experience by making the lending process faster and more accessible.
What are some of the areas in which you are using GenAI? What impact does it have?
The BFSI sector is in the early stages of exploring the potential of generative AI to enhance our customer relationship and servicing modules. Currently, there is a high focus on leveraging AI-powered chatbots and virtual assistants to streamline customer interactions. By incorporating these AI tools, NBFCs aim to provide more efficient, personalised, and timely responses to their customers’ needs. This not only improves service delivery but also allows them to scale our support operations, ensuring their clients experience smoother communication and quicker resolutions. As we continue to innovate, we foresee generative AI playing an increasingly significant role in optimising our customer engagement processes.
We plan to use AI to revolutionise its credit underwriting and loan disbursement processes. By integrating AI in our platform, we automated credit decisions, enhanced accuracy, and improved risk management. This AI system analyses extensive data to assess creditworthiness more efficiently than traditional methods. In loan disbursement, AI streamlines the process, ensuring faster approvals and fund transfers. It also allows us to handle a higher volume of applications while providing personalised loan products tailored to individual customer needs. This tech-driven approach not only boosts operational efficiency but also aligns with our goal of becoming a leading small business financing institution.
How do you see the traction for GenAI in the industry? Please highlight the key areas.
GenAI is gaining remarkable traction across industries due to its ability to produce human-quality text, images, and even code, capturing the attention of both businesses and consumers.
Several key areas are seeing the most notable impact:
Content creation: GenAI is revolutionising how content is produced, from generating articles, blog posts, and marketing copy to creating more artistic outputs like scripts and poetry. We are also exploring its use in the Software Development Life Cycle (SDLC) for creating Business Requirement Documents (BRDs) and designing test cases.
Product development: GenAI assists in the design, prototyping, and testing of new products, streamlining the entire development process.
Customer service: AI-powered chatbots and virtual assistants are transforming the customer experience by delivering more personalised, efficient support.
Research and development: AI accelerates R&D efforts by analysing large datasets and generating new, actionable insights.
Learning and development: In education, AI is reshaping learning experiences through AI-powered tutoring tools and personalised learning platforms.
Though GenAI is still in its early stages, the potential applications are vast, and as the technology continues to evolve, we anticipate even more innovative use cases that will transform how businesses operate and serve their customers.
With the increasing importance of cybersecurity, what measures are you taking to ensure the security and integrity of your IT systems and sensitive data?
As an NBFC, ensuring the security and integrity of IT systems and sensitive data is a top priority. We have implemented comprehensive measures to protect customer information and safeguard our infrastructure. Data encryption, both in transit and at rest, is employed to secure all customer information. We enforce Multi-Factor Authentication (MFA) and strict access controls to prevent unauthorised access. Our Secure Software Development Life Cycle (SDLC) ensures that security is embedded in every stage of software development and patching processes.
We also implement role-based access control (RBAC) and least-privilege principles for both employees and third-party vendors. Regular Vulnerability Assessments and Penetration Testing (VAPT) help identify and fix potential vulnerabilities, while Data Loss Prevention (DLP) solutions prevent unauthorised data access or sharing. To further bolster endpoint and server security, we use advanced Endpoint Detection and Response (EDR) systems. Additionally, we focus on employee awareness and training to combat phishing attacks, and our 24/7 Security Operations Center (SOC) provides continuous monitoring to proactively address potential threats. These initiatives collectively ensure robust cybersecurity for our operations and customer data.
Can you share some of the key challenges you faced while implementing AI/ML solutions, and how you overcame them?
Implementing AI/ML solutions comes with a unique set of challenges that require careful consideration and strategic planning. One major hurdle is data quality and availability, as ensuring high-quality, representative data is critical to avoid biased models and unreliable predictions. Integration with existing systems is another challenge, especially when dealing with legacy infrastructure. Organisations should leverage an API-based architecture to facilitate seamless integration of AI/ML solutions into existing workflows.
The shortage of skilled AI/ML professionals presents a challenge in talent acquisition. To overcome this, companies can use AI-as-a-service platforms and invest in upskilling and training their existing workforce. Additionally, scalability is crucial as solutions must grow with increasing data and user demands. Companies can utilise a cloud-based, serverless architecture to ensure scalability without sacrificing performance.
Ethical considerations are a priority, as addressing biases in data and ensuring fairness is crucial for responsible AI deployment. Regular audits of both data points and outcomes to maintain ethical standards help overcome this.
Demonstrating ROI and gaining stakeholder buy-in is another challenge, as AI/ML implementation requires significant investment. Companies are advised to start with pilot projects to showcase value, establish KPIs, and communicate success metrics regularly to stakeholders.
Performance monitoring and maintenance are critical for ensuring model accuracy over time. Setting up automated monitoring systems and establishing a regular retraining schedule to maintain model effectiveness is the way to go.
Navigating regulatory compliance is also a key consideration, especially regarding data privacy and security. Building an AI model in-house to score customer credit profiles and implementing a robust data security structure to comply with regulations are perfect ways in navigating regulatory compliance.
Indeed, overcoming these challenges requires a combination of strategic planning, effective communication, and ongoing education to ensure smooth implementation across the board.
In the next six months to one year, what are some of the new digital or technology initiatives planned?
In the coming months, one of our key digital initiatives is to leverage the OCEN (Open Credit Enablement Network) framework to build robust in-house capabilities. This will allow us to better serve and cater to our partners and alliances by streamlining credit access and enabling seamless integration with our systems. By adopting the OCEN framework, we aim to enhance our collaboration with fintechs and other lending platforms, making credit more accessible to underserved MSMEs and driving greater efficiency in our lending processes.
In the upcoming six months, we are set to enhance its credit underwriting and loan disbursement processes through advanced AI technologies. They plan to enable AI in our existing platform to automate and improve the accuracy of credit decisions. This platform will help us analyse extensive data to assess creditworthiness more efficiently, thereby reducing manual efforts and enhancing risk management. For loan disbursement, we aim to streamline the process using AI, ensuring quicker approvals and fund transfers. This approach will enable them to handle a higher volume of loan applications while providing personalised loan products tailored to individual customer needs.