Axis Mutual Fund - the mutual fund arm of one of India's largest private banks set out to bring their investor services directly to where their customers already are: WhatsApp. With nearly 9 million active investor accounts across more than 100 cities, the client wanted a channel that would make routine account queries instant, accessible, and intelligent. Beyond self-service, they envisioned connecting investors seamlessly to live customer care agents through the same familiar interface. The goal was not just convenience - it was a step toward becoming one of the most digitally responsive asset management brands in India.
Delivering a secure, scalable, and responsive WhatsApp support platform for millions of investors was far from a simple integration task. The client's environment came with strict infrastructure requirements, complex on-premise system dependencies, and performance expectations that ruled out conventional approaches.
The system had to be hosted entirely within the client's internal infrastructure. Cloud-hosted or third-party-managed solutions were not an option, placing significant constraints on deployment architecture and DevOps tooling.
Investor account data resided across multiple internal systems with no existing API layer designed for real-time messaging use cases. Securely connecting the WhatsApp platform to these systems without compromising data integrity or exposing sensitive financial records required careful API design and rigorous security controls.
With millions of active investors as a potential user base, response time was non-negotiable. The architecture needed to sustain high concurrency without degradation and needed to scale up and down elastically based on demand, rather than provisioning for peak load permanently.
A significant volume of investor inquiries follow predictable, repetitive patterns. The platform needed machine learning capabilities to recognize and resolve frequently asked questions automatically, reducing the burden on human agents while improving response speed.
Given the financial nature of the platform, any service degradation had direct customer impact. The team identified risks around queue poisoning, cache overflow, single-server scalability limits, and the absence of backup and failover mechanisms - all of which needed to be addressed before go-live.
Focaloid designed and delivered a production-grade, microservices-based WhatsApp customer support platform fully hosted within the client's secure infrastructure that combines intelligent automation, event-driven messaging, and distributed caching to deliver fast, reliable, and scalable investor experiences.
Each functional domain investor query handling, agent routing, authentication, notifications was built as an independent service exposing REST APIs. This modular structure enabled independent deployment, isolated failure containment, and targeted scaling of high-demand services without touching the rest of the platform.
All services were containerized using Docker and orchestrated via Kubernetes, enabling automated scaling, self-healing deployments, and consistent environment parity across development, staging, and production. This also directly addressed the recommended infrastructure guidance from the technical discovery phase.
Rather than synchronous service-to-service calls which introduce latency chains and tight coupling - inter-service communication was implemented using AMQP via RabbitMQ. A dedicated queue monitoring system was also implemented to detect service faults and prevent queue poisoning, one of the key risk factors identified upfront.
Redis was deployed in cluster mode to provide a distributed cache layer, dramatically reducing database round trips for high-frequency read operations. Cache expiry policies were enforced to prevent cache bloating - another risk flagged during discovery and Redis clustering ensured horizontal scalability without a single point of failure.
A machine learning layer was integrated to recognize and automatically resolve frequently asked investor questions. The engine learns from interaction patterns over time, continuously improving resolution rates and reducing escalation to live agents.
Jenkins pipelines were implemented to automate build, test, and deployment processes, enabling rapid and consistent delivery of enhancements without manual release overhead. All activities and messages were logged end-to-end for audit, compliance, and operational visibility.
We followed a structured discovery-to-delivery methodology, ensuring risks were surfaced and mitigated before any code went to production.
Ran structured discovery workshops with the client's business and technology stakeholders to gather both functional and non-functional requirements. Produced a comprehensive requirements document covering architectural constraints, integration needs, security requirements, performance benchmarks, and compliance expectations.
Conducted a technical analysis to identify the major risk factors - infrastructure constraints, queue poisoning, cache overflow, failover gaps, and security dependencies. Translated findings into an architecture blueprint covering microservices topology, Kubernetes configuration, message broker design, cache strategy, and API security patterns.
Delivered the backend microservices in Node.js using a Reactive design pattern, with each service exposing REST APIs for consumption by the API gateway and peer services. Angular was used for the agent-facing frontend. RabbitMQ and Redis were integrated as part of the performance and resilience layer.
Integrated the machine learning FAQ engine and executed rigorous stress testing to identify and resolve bottlenecks before production cutover. QA coverage encompassed functional testing, integration testing, load testing, and security validation.
Implemented Jenkins CI/CD pipelines, finalized Kubernetes deployment configurations, enforced cluster mode for Redis and RabbitMQ, set cache expiry policies, and validated the backup and failover system ensuring the platform was production-ready and operationally resilient from day one.
For a mutual fund house managing assets for nearly 9 million investors across India, the ability to deliver instant, accurate, and secure service at scale is not a differentiator - it is a baseline expectation. WhatsApp has become the primary communication channel for a significant share of India's financial consumers, and meeting investors where they are is increasingly a competitive necessity. By building a microservices platform that is independently scalable, operationally resilient, and ML-augmented, Focaloid helped the client make a structural leap not just adding a new channel, but engineering a foundation capable of evolving with the business. As investor volumes grow and new service use cases emerge, the modular architecture means each capability can be enhanced, scaled, or replaced without disrupting the whole. This is the kind of platform that compounds in value over time.
Whether you're looking to add new digital channels, modernize a monolithic backend, or scale an existing platform to millions of users?
We bring the architecture expertise to do it securely and without disruption. We help financial services companies turn complex infrastructure challenges into fast, reliable, and scalable customer experiences.