Powered By Inc42 Brandlabs

Brandlabs

The brand solutions arm of Inc42 Media combining Inc42’s creative and editorial strengths to create compelling stories for brands partnering with it.

How Enterprises Are Building Data Platforms That Never Slow Down

Plus-New-Logo-300x63-2.webp

Become A Startup Insider With Inc42 Plus

Join our exclusive community of business leaders &makers for in-depth tech stories and intelligence on India’s tech economy you won’t find elsewhere.

Become a member

Unlimited Access

Unlock unlimited access to 500+ monthly new stories and Inc42’s complete archive of 40,000+ stories.

Become a member

Inc42 Academy

Exclusive, in-depth and actionable sessions from 300+ industry leaders, founders & operators to help you level up.

Become a member

Startup Deals

Over $250,000 worth deals on software & tools to help your business scale up.

Become a member

Member-Only Events

Connect with editors and industry peers through our members-only community

Become a member

Clutter-Free Experience

Enjoy an ad-free and clutter-free experience as a member, focus on what really matters to you

Become a member

Support Independent Journalism

Your contribution will go on to support and empower our unbiased & data-driven journalism

Plus-New-Logo-300x63-2.webp

Become A Startup Insider With Inc42 Plus

Join our exclusive community of business leaders &makers for in-depth tech stories and intelligence on India’s tech economy you won’t find elsewhere.

Choose Your Plan

2 YEAR PLAN
₹19999
₹5999
₹249/Month
Cancel Anytime
1 YEAR PLAN
₹9999
₹3499
₹291/Month
Cancel Anytime
Cancel Anytime

Join Us In Celebrating 5 Years Of Inc42 Plus!

Unlock special offers and join 10,000+ founders, investors & operators staying ahead in India’s startup economy.

2 YEAR PLAN
₹19999
₹5999
₹249/Month
UNLOCK 70% OFF
Cancel Anytime
1 YEAR PLAN
₹9999
₹3499
₹291/Month
UNLOCK 65% OFF
Cancel Anytime

In the tense days leading up to Hurricane Frances in 2004, as Florida braced for impact, Walmart’s data team uncovered something surprising. While it was no shock that flashlights and batteries flew off the shelves, the data told a more unexpected story. The sales of strawberry Pop-Tarts were surging. Not just slightly, but dramatically.

Digging deeper, the team realised that whenever a hurricane loomed large, strawberry Pop-Tarts became a top-selling item. Armed with this unlikely insight, Walmart quickly adapted. Stores in hurricane-prone areas were stocked not only with emergency supplies but also with pallets of Pop-Tarts.

The result was an incredible boost in sales and a powerful reminder of how data, even in the most unpredictable moments, can uncover patterns that intuition alone might miss.

With AI reshaping the business world, the demand for real-time, data-driven decision-making has soared further. Enterprises are now striving to engineer scalable, intelligent data platforms that deliver faster, more precise insights at scale.

The numbers alone tell a compelling story. According to Fortune Business Insights, the global data analytics market size was valued at $64.99 Bn in 2024. The market is projected to grow from $82.23 Bn in 2025 to $402.70 Bn by 2032, exhibiting a CAGR of 25.5% during the forecast period. 

According to Feedzai, 90% of financial institutions use AI to expedite fraud investigations and detect new tactics in real time. Wells Fargo, a multinational financial services company, processes over 50 Mn transactions daily through their fraud monitoring infrastructure, analysing 100% of transactions in real-time with machine learning analytics. Its system maintains response times under 10 milliseconds per operation — a feat impossible with traditional database architectures. 

Meanwhile, companies using real-time analytics see a 25% increase in customer retention rates within six months. However, 76% of practitioners cite siloed or fragmented data as a major obstacle to real-time personalisation.

This means that with GenAI, Agentic AI, and large-scale inference models coming to the fore, building a robust data infrastructure is no longer a technical option but a strategic necessity. To explore this, Inc42 and Couchbase recently hosted a virtual panel discussion on ‘Building Scalable Data Platforms That Never Slow Down’, as part of an ongoing series titled ‘Boardroom: The AI & Data Playbook’, featuring a lineup of seasoned technology leaders:

  • Satya Kaliki, CTO, Infra.Market
  • Sandeep Varma, Head – Data Science & Engineering, PhysicsWallah
  • Prakhar Verma, Chief Software Architect​, Capillary​ Tech
  • Shashidhar Ramakrishnaiah, CTO, Fractal
  • Krishna T, Regional Business Head, Couchbase

Moderated by Sameer Dhanrajani, founding partner at MIRAI Ventures and CEO of 3AI, the conversation unpacked the practical challenges, architectural debates, and governance imperatives behind building enterprise-grade data platforms.

How Indian Enterprises Are Rethinking Data Platforms For AI At Scale

Sameer kicked off the session by underscoring how decision-making at scale, powered by AI, is fast becoming central to boardroom agendas. But achieving this in real time requires a strong foundation – a resilient, scalable data platform.

Sandeep reflected on PhysicsWallah’s journey, which evolved from making business decisions on Google Sheets to deploying a robust lakehouse architecture. “When we scaled from a test prep business to 30 verticals across online and offline modes, our biggest challenge was optimising for real-time needs without incurring unsustainable costs,” he explained.

Shashidhar pointed to a core challenge – the constant push-and-pull between centralised infrastructure teams and market-facing business units. “The markets want to move fast, often faster than central IT. Data mesh and fabric concepts are helping, but it’s the advent of Agentic AI that’s enabling foundational capabilities to scale with agility,” he said.

Adding to this, Krishna delved into the challenge of data sprawl. “Most enterprises are dealing with fragmented data systems. This makes it difficult to deliver consistent, real-time insights across business functions. It’s not just about speed but precision.”

Prakhar explained that building a multi-tenant platform to serve various enterprise clients requires high customisability and adaptability.

“Different businesses have different KPIs and metric needs. While real-time matters for some, near real-time is sufficient for others. There’s no one-size-fits-all,” he said.

When asked if platforms must be built brick-by-brick or can be accelerated with modern tools, Satya pointed out, “There’s no need to start from scratch anymore. Metadata platforms are mature. Unless you’re building platforms yourself, use what’s already out there to move faster.”

Concluding the discussion, Sameer said that what’s true today in AI and data might be irrelevant in 3 months due to the fast pace of changes in the segment. 

But as the panel showed, the path to building effective, scalable, and intelligent data platforms is becoming clearer — with agility, governance, and context-awareness leading the way.