Srujani Elango: Advancing Data Engineering Standards From Research Labs to Robotics Reality

In Amazon Robotics, many autonomous systems require precision for various processes. Data quality must be prioritized and baked into the operational infrastructure. When this isn’t done, blockages and delays will happen, affecting entire crucial systems.

“This is exactly why translating rigorous academic research on streaming data resilience into production environments has become essential for handling the heaviest workloads without compromising reliability,” says Srujani Elango, a Data Engineer specializing in robotics at Amazon.

Her work illustrates how foundational research in data quality can directly reduce downtime, accelerate decision-making, and lower costs in one of the most demanding industrial settings.

Elango is recognized as an expert with a passion for her work. She shares her fascination: “What drew me to data engineering was realizing that data science and analysis only deliver value when the underlying systems are trustworthy at production scale. The engineering layer is where intent becomes a reliable outcome.”

And that’s not all – Elango is known for advancing data engineering standards in every field she’s been in so far.

An Analytical Background

Elango was first and foremost a dreamer and a student. She was born in India, and growing up, had a natural affinity for mathematics and computer science. Elango secured a high rank in her state in both areas, with 97% as her grade percentage.

Eventually, she was drawn to Anna University in Chennai, where she earned a Bachelor of Science in Information Technology: “I decided to focus on analytics rather than pure computer science, as I felt it was the perfect marriage. It also fascinated me to see the connection and relationship of analytics. I see analytics not as a narrow technical skill but as a discipline that reveals relationships across systems, processes, and business realities”

Elango’s approach is grounded in her background, which holds a strong technical foundation. From that initial spark of interest, she now holds a Master of Science in Information Systems from Northeastern University, with coursework in engineering big data systems and advances in data science. Upon broadening her horizons through her degrees, Elango wanted to learn and utilize her skills further: “During my undergrad, I realized I wanted to go deeper,” she recalls.

“A friend’s brother was pursuing a master’s at Carnegie Mellon University and working on an exciting market basket analysis project, where he studied customer purchasing patterns, such as which products were often bought together at a supermarket, to help businesses improve recommendations and decision-making. That sparked my interest in advanced data applications.”

Later on, her roles at AWS, where she built ETL pipelines serving more than 4,000 users and collaborated on machine-learning enrichment of customer datasets, as well as at DraftKings, provided breadth across cloud-scale data platforms and high-velocity transactional environments.

That range of experience informs her current emphasis on systems that remain reliable when schemas shift, volumes spike, and downstream consumers multiply. She handles such projects with care and an innovative approach: “AWS work enriches customer datasets for better personalization, benefiting businesses from small shops to businesses of varying sizes, including well-known global enterprise customers.”

Soon enough, Elango transformed her entire career. Data engineering became her passion, and she was fascinated by how it served as the foundation for actionable business insights. She was drawn to building robust data systems, which defined her future learning and career.

She reflects on these interests and says: “These three fields of data science, analysis, and engineering overlap and must work in harmony to drive success. That realization propelled me into data engineering, where I now specialize in powering robotics solutions at Amazon.”

A Commitment to Research

One thing’s for certain: the research Elango goes through is not just theoretical. As she says: “The replay-safe framework from our research didn’t stay in the paper. We applied the same principles to live event streams so that schema evolution no longer forced teams to choose between completeness and recoverability.”

One of the things Elango is best known for is her paper, “Resilient Streaming: A Replay-Safe Data Quality Framework for Evolving IoT Schemas,” presented at the ICDBA conference.

The paper addressed a persistent problem in IoT and event-driven pipelines. Schemas change over time as systems evolve, but downstream applications and analytics still require reliable historical replays. With Elango’s work, a framework was developed to introduce replay-safe mechanisms that preserve data integrity through schema evolution, eliminating the painful trade-offs among freshness, completeness, and recoverability.

She shares: “My co-authored research on resilient streaming data quality for evolving IoT schemas (improving completeness to 99.8% and slashing recovery times) offers frameworks applicable to robotics, smart grids, and beyond.”

Elango is interested in more than just the basics of her work; she dives headfirst into research. As a curious person by nature, her work is driven by the desire to innovate in order to raise the quality of engineering: “I want to be known as the curious, reliable engineer who learns continuously and delivers results, no matter the challenge. Someone that others can be entrusted with complex problems, knowing they’ll be solved effectively.”

In Elango’s research on resilient streaming, she focused on systems that can retrieve, store, and process high-volume data while maintaining reliability as schemas and source systems evolve.

This has clear relevance to robotics, where machines and operational platforms continuously

generate data on movement, performance, errors, delays, and task status. Her work emphasizes the importance of building pipelines that preserve data quality and continuity, so teams can trust the information flowing into analytics platforms and use it to make timely operational decisions.

Elango says: “Even with the rise of AI, high-quality data remains its essential fuel. There’s constant evolution, which includes new tools, larger scales, and emerging challenges. Thus we have the opportunity to learn and pivot in a multitude of ways.” Even with this constantly changing process, she is relentless, using the changes as a source of excitement in approaching her work rather than as barriers to her learning curve. She continues: “This sustainability and scalability make the field truly exciting.”

That research is required in the ever-evolving process of robotics and design, as Elango shares: “In robotics fulfillment, data isn’t just infrastructure. It’s the real-time nervous system. When schemas shift, or upstream systems evolve, you need mechanisms that protect both freshness and the ability to reliably replay history. That was the core problem we set out to solve.”

And solve it she did, despite the challenges it offered up, setting a new benchmark in the arena of data engineering and robotics. Aside from the amount of information she has to process, Elango must also consider the system's reliability.

And what distinguishes her methodology is its ability to maintain high-quality, real-time data while accommodating evolving data schemas and preserving historical records. Her approach builds resilience directly into the architecture. Robotics systems adapt to change without sacrificing accuracy or operational continuity.

As Elango points out: “Processing six terabytes daily across more than a thousand workflows sounds like a scale problem, but the harder challenge is keeping every pipeline observable and recoverable. When a delay hits robotics operations, the cost cascades immediately.”

In her experience at Amazon Robotics since 2022, one can clearly see Elango’s discipline in creating observable data flows despite massive scales. She designed and implemented a real-time streaming platform that ingests operational events from ERP and PLM applications, and five additional APIs. Events stream into an AWS data lake via Kafka and EMR/Spark, organized through a Bronze/Silver/Gold medallion architecture.

The architecture standardizes ingestion while enabling progressively refined analytics layers. The outcome has been concrete: a 30% reduction in operational downtime and roughly

$600,000 in annual savings for the robotics business unit.

Discussing her crucial role in the creation of the architecture, Elango says: “The Bronze/Silver/Gold medallion architecture gave us a disciplined way to land raw events from ERP and PLM applications, and the APIs into the data lake, then progressively refine them. It removed the ad-hoc fixes that used to create fragile dependencies across teams.”

Beyond this, Elango also worked on the Data Lineage Platform, which integrated metadata from multiple sources in Open Lineage format. Then, it was visualized in DataHub, after which Elango built a scalable architecture using AWS services including Airflow, Glue, Lambda, Spark, Kafka, and EKS.

Elango reflects on the project's success, noting how different it is now from the previous workflow: “Before we built the lineage platform, tracing an anomaly back through five different enterprise applications could consume hours of engineering time,” she recalls.

“But integrating metadata in Open Lineage format and surfacing it in Datahub changed the economics of debugging and compliance work.”

The result? The company spent less time on debugging and compliance, and more time on other productive, innovative solutions. And Elango’s work proved adept at improving traceability by 95%, accelerating debugging, and enhancing compliance.

The added benefits to her team were that they could enjoy deeper insights into robot performance, warehouse operations, and delivery timelines.

She says: “Because the system was designed to be reusable among Amazon organizations, like Amazon Business, it made a major impact and had effective internal use. The other result was we had excellent voice-of-customer feedback.”

Elango has also led the development of scalable ETL pipelines that can now process six terabytes of data daily across more than 1,000 workflows supporting Amazon Robotics' supply chain and manufacturing teams. These pipelines deliver the near-real-time visibility required for robotics delivery and fulfillment operations.

She remarks: “It was a unified ingestion layer for heterogeneous data streams (5-6+ sources) processing 6 terabytes daily. This real-time solution fed into a modern lakehouse architecture, ensuring reliability across 1,000+ workflows. It prevents business disruptions and supports scalable insights, aligning with current data engineering best practices.”

Another additional project under Elango’s leadership is an AI-enabled analytics chatbot that fuses analytical and historical operational datasets: “The bot was about removing friction between question and answer,” she says.

The work meant the operations teams and leadership could query the system for timely insights, reducing the latency between question and decision in manufacturing workflows: “Now the fused historical and live datasets can be accessed directly, which compresses the cycle from insight to decision on the manufacturing floor,” Elango explains.

“It just makes everything run a lot more smoothly and saves time and effort for all concerned. And every architectural choice was validated by outcomes the business could feel. For example, there was thirty percent less operational downtime and hundreds of thousands in annual savings. Those numbers matter because they reflect real improvements for the teams running fulfillment every day.”

Elango has also worked to raise the visibility and standards of data engineering itself. She has mentored through CodePath and participated in STEM outreach programs for students, helping to broaden the talent pipeline entering the field.

Shaping the Future of Robotics

More than an efficient worker, Elango is also a skilled mentor. She wants to invest heavily in the skills of future engineers, and says: “Talent pipelines matter as much as data pipelines. Through mentoring and STEM outreach, I try to show emerging engineers that data engineering isn’t just plumbing; it’s the discipline that makes large-scale automation trustworthy and sustainable.”

The broader stakes are significant. Robotics, AI-driven automation, and smart infrastructure all depend on data systems that can ingest, transform, and serve information with high fidelity even as upstream sources change.

Elango’s combination of research insight and production execution offers a practical model for building those systems. By insisting on scalable infrastructure, rigorous lineage and observability, and measurable reliability improvements, she is helping define what

“production-grade” data engineering looks like for environments where failure is expensive and real-time insight is a competitive advantage.

Despite all the challenges she faces in her day-to-day work, Elango is enthusiastic about the future of robotics, and she’s willing to keep building towards it: “As robotics and intelligent automation expand, the bar for data engineering keeps rising. The systems we build today must remain reliable even as upstream sources, volumes, and downstream consumers continue to change. This is exactly why resilience and observability can’t be afterthoughts.”

This kind of discipline and responsibility is something Elango is known for. Peers have taken notice, including Aswath Kirubakaran, a Senior data engineer at Meta Reality Labs. It is a dedicated research and development division within Meta Platforms (formerly Facebook, Inc.) focused on building the hardware, software, and underlying technologies for virtual reality (VR), augmented reality (AR), and wearable computing.

Kirubakaran got to know Elango through the Northeastern University data engineering community and has worked alongside her in the industry ever since.

He says: “I've noticed that Srujani takes ownership of whatever she's responsible for. She doesn't settle for doing only what is expected. Instead, she looks for opportunities to improve existing systems and make them more reliable for everyone who depends on them. That mindset is reflected in the impact her projects have had over the years.”

Another colleague is Guhan Kumaresan, Senior Data Analyst at Yubico, a cybersecurity company best known for inventing the YubiKey, a hardware authentication device used to protect computer systems, networks, and online accounts from unauthorized access.

As another data engineer, Kumaresan has professionally known Elango for over eight years. He notes: “She consistently focuses on building reliable, scalable solutions rather than quick fixes and takes ownership of complex technical problems until they are resolved. Her dedication to quality, attention to detail, and a desire to continuously improve both herself and the systems she builds are evident throughout.”

For Elango, success seems inevitable. She shares: “I’m wired to solve problems and build solutions. From a young age, this drive has defined me. I thrive on challenges, adaptation, and growth in a non-linear career. Making systems better and contributing to transformative robotics keeps me motivated daily.”

Elango’s biggest desire in her work is to create stronger industry standards, as she advocates for greater visibility and higher-quality work. She remains curious and relentless in her drive to learn and deliver results. She highlights this importance: “Data engineering often operates in the background but is foundational. Establishing consistent best practices across sectors would elevate the field and recognize its critical role in robotics and beyond.”

The next iterations of robotics and data engineering will no doubt have their challenges, but Elango is here to consistently rise above them.

About the Author

Sashindra Suresh is an experienced writer specializing in artificial intelligence, software development, and emerging technologies. With a strong ability to translate complex technical concepts into clear, engaging insights, she has contributed to a wide range of publications and platforms. Her work focuses on making cutting-edge innovations accessible to both industry professionals and curious readers alike.