As artificial intelligence reshapes the capital markets industry, Mirae Asset Sharekhan is building a multi-layered AI strategy spanning customer engagement, research intelligence, operational automation, cybersecurity and developer productivity. The company’s AI journey began around late 2024, following organisational consolidation efforts after the broader group-level integration phase. Since then, the company has focused on building a foundational data architecture. A key pillar of this strategy has been the creation of a centralised data lake aimed at breaking operational data silos and enabling structured as well as unstructured enterprise data to be consolidated into a single intelligence layer. At the leadership level, AI has emerged as a strategic priority, with the board and senior management actively driving the creation of an enterprise-wide AI roadmap.
The company said it has already started witnessing early operational gains from several AI-led initiatives. Multiple internal processes that previously took hours are now being completed within minutes. Within technology operations, AI is helping development teams reduce manual effort, accelerate debugging and identify performance bottlenecks more efficiently.
Mirae Asset Sharekhan is approaching AI with a broader focus on customer excellence, operational intelligence and risk management, while also leveraging the wider AI capabilities and experiences available across the global Mirae Asset group ecosystem.
At the group level, AI has emerged as one of the core strategic priorities. The group has also expanded its capabilities through acquisitions such as Wealthspot, Mirae Asset’s AI-focused investment and financial intelligence platform based in New York and Stockspot, an online robo-advisory and digital wealth-management platform in Australia. These experiences and learnings are now being leveraged across businesses, including Mirae Asset Sharekhan.
AI to Enhance Customer Engagement
One of the primary focus areas for Sharekhan is understanding customer behaviour more effectively through the intelligent use of data. The company is exploring how it can analyse customer portfolios, identify behavioural and investment patterns, and create more personalised engagement models based on customer preferences and investment interests. The larger objective is to strengthen customer engagement while enabling Sharekhan to deliver more relevant value-added services.
The company is also looking at how AI can enhance the utilisation of its research capabilities. “Mirae Asset Sharekhan has transformed how customers consume our research. Clients can now access our expert-backed research calls simply by prompting on leading AI platforms such as Claude, powered by our AI MCP layer that connects these platforms to Mirae Asset Sharekhan’s research ecosystem—turning deep research into quick, actionable insights. Once authenticated with their credentials, customers can prompt their AI platforms to bring together their portfolio data and Mirae Asset Sharekhan’s research perspectives and relevant market and news updates. This kind of analysis by customers using the right prompts allows them to extract unique information relevant to them,” said Madhusudan Warrier, Chief Technology Officer (CTO), Mirae Asset Sharekhan.
Another major area of focus is relationship manager (RM) enablement. Traditionally, Sharekhan has operated with a strong branch-led presence, supported by regional managers and relationship managers handling dedicated customer portfolios. The company is now exploring how AI-driven insights can help these managers improve customer engagement quality while also increasing their servicing capacity.
The idea is to provide relationship managers with more contextual and actionable customer insights before interactions take place. This enables conversations to become more targeted towards what the customer is likely to be interested in, thereby improving the quality and relevance of engagement. This can significantly improve RM productivity and allow managers to efficiently handle a much larger customer base over time.
Information Security and Risk Management
From an information security perspective, the company is leveraging AI to analyse large volumes of security logs generated within its infrastructure. “The objective is to identify trends, detect potential attacks and generate actionable security insights for the security operations centre (SOC). AI can significantly improve visibility into evolving cybersecurity threats and strengthen its overall security monitoring capabilities,” said Warrier.
AI is also being explored for risk monitoring and anomaly detection across customer trading behaviour. By analysing trading patterns, the system can potentially identify unusual or suspicious activities and proactively alert the company’s risk and operations teams. The broader intention is to build an AI-enabled intelligence layer across the organisation’s ecosystem to improve surveillance and operational responsiveness.
From Hours to Minutes
On the operations side, the company is focusing on automating several post-market processes that are traditionally manual and time-intensive. In broking operations, a significant portion of operational work begins after market hours, including activities such as processing contract notes, reconciliations and end-of-day operational workflows. “The company is evaluating how AI can automate and streamline many of these activities, reducing dependency on manual intervention and minimising the possibility of human error.
Several processes that previously took hours are now being completed within minutes,” said Warrier. Operational teams that earlier had to manually scan and analyse large volumes of data are increasingly being supported by AI-driven systems that can read, interpret and present information through simplified visual interfaces. Teams can now identify issues more efficiently using intuitive indicators such as colour-based alerts, allowing them to focus attention only on areas requiring intervention. According to Warrier, this is fundamentally changing the way operational teams work and is also prompting a broader re-evaluation of existing workflows and operating models.
Overall, Warrier stated, Mirae Asset Sharekhan is currently experimenting with multiple AI-led initiatives across customer engagement, research, cybersecurity, operations and intelligence systems. However, the central objective remains consistent: every AI deployment must ultimately create meaningful value either for external customers or for internal users. Warrier said, if an AI solution does not materially improve customer experience, operational efficiency or decision-making, its long-term value remains limited.
RoI on AI
At present, the company is focusing more on defining measurable operational and productivity benchmarks rather than immediately quantifying direct financial returns. “One of the key internal targets is to significantly reduce software development timelines without proportionately increasing headcount. For example, the company is aiming to reduce development turnaround times by nearly 50% through the use of AI-assisted development practices,” stated Warrier.
Similarly, several operational processes are being evaluated with the objective of materially reducing execution timelines through automation and AI-led intelligence layers. These are among the initial performance parameters being tracked as part of its broader AI adoption strategy.
On the software engineering side, Warrier said, “AI-generated code currently accounts for roughly 10–15% of its overall coding activity,” However, this phase is still largely focused on experimentation, learning and understanding how AI can be effectively embedded into broader technology workflows and engineering practices.
The company expects AI-assisted development adoption to increase significantly over the coming months as teams gain greater confidence from ongoing implementations and observed productivity gains. “Internally, the company has set a target of increasing the share of AI-generated or AI-assisted coding activity to at least 25% before the end of the year.”
Overall, the current approach is centred on gradual experimentation, replication of successful use cases and operational learning, before moving towards large-scale ROI measurement and broader AI-driven transformation initiatives.
