Hina Gandhi, Senior Software Engineer, brings nearly a decade of experience from industry leaders like Cisco, VMware, and CloudHealth Technologies. In this interview, Hina shares lessons from scaling distributed systems, leading the transition from monoliths to microservices, and navigating the evolution of engineering leadership in a hybrid world. She also highlights common pitfalls in SaaS development and explores the future of cloud computing shaped by AI. Read on for a candid look into the strategies and challenges shaping the next generation of cloud innovation.
Explore more interviews here: Srinivas Sandiri, Technology Leader in Digital Transformation — The Evolution of AI in CX, Automation vs. Human Touch, Ethical AI, Cross-Team Alignment, and Preparing the Next Generation
Your career spans some of the most innovative companies in the tech industry—Cisco, VMware, and CloudHealth Technologies. What are some defining moments that shaped your expertise in distributed systems and cloud computing?
I began my journey in distributed systems and cloud computing at CloudHealth Technologies, first as an intern and later as a full-time engineer. As one of the early engineers, I played a key role in designing and developing Microsoft Azure cloud cost management features, which became a major product offering and helped capture new market share. This experience gave me hands-on exposure to distributed systems and reinforced my understanding of how essential they are for high availability and scalability in cloud-based applications.
During my time at CloudHealth (later acquired by VMware), I worked on optimizing critical data processing jobs used for cloud insights reporting. The existing job was a major performance bottleneck—despite running on AWS’s highest-tier EC2 instance, it took over 24 hours to complete, negatively impacting the user experience. To resolve this, I designed and implemented a distributed data processing system that divided a single large job into multiple parallel tasks, drastically improving performance by 10x, reducing costs, and making the system significantly more scalable. This project deepened my expertise in distributed data processing and real-world cloud optimization strategies.
At VMware, working on large-scale applications further refined my skills, setting the stage for my next challenge at Cisco. There, I led an end-to-end transformation of a monolithic application into a cloud-based microservice. The original system, responsible for exporting customer vulnerability data from on-premise assets and software applications, faced severe performance issues due to noisy neighbor problems and resource contention. I redesigned the system using AWS services in a distributed architecture, improving performance by 5x, reducing costs, and enabling on-demand resource allocation to handle unpredictable workloads efficiently.
Across my 9+ years in the industry, I have not only deepened my expertise in cloud computing and distributed systems but have actively applied it to solve scalability and performance challenges in real-world applications.
---
Many organizations struggle with the transition from monolithic to microservices architectures. Can you walk us through a real-world transformation you led, the biggest hurdles you faced, and how you measured success?
In my 9+ years of experience, two of the most impactful projects which were transitioned from monolith to microservice were at VMware and Cisco. In both cases, the monolithic applications faced severe performance issues, with job processing times exceeding 24 hours, leading to service outages and negatively affecting user experience. To address these challenges, transitioning to microservices was essential, as it allowed us to break down large, tightly coupled jobs into smaller, independent services that enabled parallel processing, better scalability, and fault isolation.
One of the biggest hurdles in this transformation was designing an architecture that not only improved performance but also remained cost-efficient. This required careful selection of technologies and system design principles, including asynchronous processing, event-driven architecture, and a database-per-service approach to reduce latency and optimize resource utilization. Another major challenge was ensuring data consistency and integrity, as any discrepancies between the old monolith and the new microservices-based system could raise concerns from customers. To mitigate this risk, an extensive 1:1 data validation was conducted, running both systems in parallel and comparing results before fully transitioning. Additionally, performance testing played a crucial role in the process, as it was needed to ensure that each component of the new service functioned optimally. This included verifying that the selected database could efficiently handle large-scale read and write operations and that auto-scaling mechanisms dynamically allocated resources based on fluctuating workloads.
Once the project was deployed to production, measuring success became critical. The most important metrics included performance improvements, cost efficiency, scalability, and fault tolerance. Compared to the old monolithic system, processing times were reduced from over 24 hours to just a few minutes, cloud resource utilization improved, resulting in approximately 30% cost savings, and the new system handled more concurrent requests without performance degradation. Furthermore, fault isolation mechanisms ensured that failures in one service did not cause system-wide outages, significantly improving overall reliability. These transformations not only enhanced system efficiency but also led to a better customer experience by enabling faster data processing, seamless scalability, and reduced operational costs.
As cloud computing continues to evolve, where do you see the biggest opportunities—and challenges—for enterprises looking to scale their distributed systems?
Cloud computing has rapidly evolved in recent years, leading to increased adoption of serverless architectures across the industry. Platforms like AWS Lambda and Google Cloud Functions help minimize operational overhead while enhancing cost efficiency. Many organizations are also embracing hybrid cloud models, allowing workloads to run both on-premises and in the cloud to balance performance and cost. However, this growing adoption brings added responsibility—particularly in cost management, where expenses can escalate due to unnoticed overuse, misconfigured autoscaling, or inefficient resource utilization. Additionally, security risks may arise from open network ports or misconfigurations that could expose sensitive data.
The “Future of Work” in tech is a hot topic. With the increasing shift towards remote and hybrid teams, how do you see engineering leadership evolving to maintain productivity, innovation, and strong team dynamics?
Hybrid work has become the new norm, and as hybrid teams continue to grow, engineering leadership must embrace flexibility—allowing team members to manage their own schedules while maintaining productivity and innovation. For fully remote teams, fostering strong team dynamics is essential. One approach is to host informal virtual gatherings, such as Friday happy hours, to encourage personal connections beyond work-related discussions. Additionally, leadership can organize in-person offsites, bringing remote team members together to align on the product roadmap, review annual OKRs, and engage in deeper conversations around customer feedback and strategic next steps.
Your work involves building scalable, high-performance applications. What are the biggest mistakes companies make when designing cloud-based SaaS solutions, and how can they avoid them?
I’ve observed companies over-engineering applications to accommodate potential future use cases that may never materialize. This often delays the delivery of scalable, high-performance solutions, forcing teams to rely on existing underperforming systems for longer than necessary. A more effective approach is to first focus on resolving current performance challenges, then iteratively evolve the solution to support future needs.
You’ve worked across different tech giants—each with its own engineering culture and technology stack. How do you adapt and innovate in different environments while ensuring long-term technical vision?
Throughout my time at various tech giants, I’ve consistently adhered to the core principles of building scalable and efficient systems. While each company faces challenges with legacy or inefficient architectures, the pressure to ship features quickly often leads to these issues being deprioritized. I believe it’s crucial to strike a balance—driving innovation to stay competitive in the market, while also modernizing and optimizing existing systems to ensure they remain robust and performant under scale.
Looking ahead, what excites you the most about the future of cloud computing and distributed systems? Are there any emerging trends or technologies that you believe will redefine the industry in the next five years?
With the rise of AI, I believe the next major trend will be the integration of AI into distributed systems. We’ll see widespread adoption of AI-driven approaches to scale resources dynamically, monitor system health in real time, leverage predictive algorithms for load balancing, and detect anomalous or malicious activity across distributed environments.
