Vinay Shyam Donakanti

Vinay Shyam Donakanti

Research Data Analyst Intern

Builds the data infrastructure that makes child-centered research possible. Transforms raw classroom observations into research-ready datasets — digitizing, cleaning, and structuring Montessori observation records so that patterns in children's learning can be seen, tested, and understood.

research@blueblocks.in

• LinkedIn

> "Children generate data that tells stories we're only beginning to learn how to read. The future of education lies in learning their language rather than forcing them to speak ours."

Researcher Profile

Vinay Shyam Donakanti builds the data infrastructure that makes child-centered research possible. As Research Data Analyst at Blue Blocks, he transforms raw classroom observations into research-ready datasets — digitizing, cleaning, and structuring Montessori observation records so that patterns in children's learning can be seen, tested, and understood.

The work changed how he thinks about data. "Every time I code a child's activity," he observes, "I'm flattening a rich, multidimensional moment into a single category. The question that haunts me is: what are we missing?" He's become obsessed with the gap between what we measure and what actually matters — what he calls the difference between "legible" and "illegible" systems.

> "How much of a child's learning happens in the spaces between our data points?"

The Researcher's Lens: The Child as Scientist

While processing classroom video data, he noticed something the original observer had missed. A child spent fifteen minutes repeatedly dropping objects from different heights. The field notes read "exploratory play." But watching the footage, Vinay saw something more systematic: the child was varying one parameter (height) while keeping others constant (object, surface), observing outcomes, forming hypotheses.

"They were conducting experiments," he realized. "Not because someone taught them the scientific method — but because that's what curiosity looks like when you let it run. Children are natural researchers not because they follow our methods, but because they embody the core of research: systematic curiosity, hypothesis testing, iteration. Our job isn't to teach them to be researchers. It's to not interfere with the researchers they already are."

The Origin of Inquiry

The insight came during his first year of engineering, stuck on an algorithm problem. After hours of frustration, he took a walk and found himself watching children in a nearby park.

"They were solving complex coordination problems — who goes first, how to share limited resources, how to modify rules when they didn't work. They were debugging their play in real-time, iterating on solutions, learning from failures without the paralysis of perfectionism that had kept me stuck."

That moment reframed everything. Learning, he realized, happens naturally when we create the right environment for inquiry. The children weren't following a curriculum. They were doing what humans do when not constrained by formal structures: experimenting, failing, adjusting, trying again.

The Uncomfortable Truth

"Most of what we call 'learning difficulties' are actually data interpretation problems — we're measuring the wrong things, or measuring the right things in the wrong ways."

A child who struggles with traditional math tests might be brilliant at spatial reasoning or pattern recognition, but our measurement systems are too crude to capture it. The problem isn't the child — it's our metrics. If we spent half the energy we put into standardizing assessments into developing richer, multi-dimensional measurement tools, we'd discover that far more children are "gifted" than we currently recognize — just not in the ways our systems are designed to see.

Current Obsession: Legible vs. Illegible Systems

His intellectual focus is the tension between measurement and meaning. Every dataset requires simplification — but what gets lost in the translation from lived experience to coded category?

"I'm obsessed with finding ways to preserve the richness while still enabling analysis," he explains. "Maybe through multi-dimensional tagging systems, narrative annotations alongside structured data, or visualization methods that show uncertainty and context. The goal is to build tools that are honest about what they don't capture — not just efficient at capturing what they do."

The Vital Mystery

If given unlimited resources, he would investigate how children develop their internal models of how the world works — and how those models evolve through experience.

"We have massive datasets on what children learn — test scores, milestones — but remarkably little on how they learn. The actual cognitive processes. The moment-by-moment sense-making. The theories they construct and revise. I'd want to track not just outcomes but the learning process itself, using observational data, think-aloud protocols, maybe even neuroscience. The goal would be to understand learning as it actually happens — not just as we've designed our systems to measure it."

At Blue Blocks

He builds the data pipelines that transform classroom observations into research-ready datasets — digitizing records, mapping activities to standardized learning categories, ensuring consistency and reproducibility across the institute's growing archive of child observation data.

But his ambition extends beyond infrastructure. "Most educational research happens in universities, far from classrooms, with data that's months or years old. Blue Blocks represents something rare: embedded research where children and researchers learn from each other in real-time. I want to build tools that practitioners can actually use — not just academic papers, but dashboards, visualizations, insights that help educators understand and support children better."

Training & Technical Skills

• B.Tech in Computer Science & Data Science — Institute of Aeronautical Engineering, Hyderabad (2022-2026)
• Data Science Certification — Corizo Pvt Ltd / Wipro (2025)
• Data Analytics Certification — Deloitte Australia / Forage (2025)
• Technical: Python, pandas, machine learning, data pipeline development, full-stack development
• Current Project: Educational data analysis and standardized coding schemas for Montessori observation data

Related

• All Team
• Governance
• Research Team