Class 12 Student | Finance Enthusiast | Decoding Markets through Code
While my younger brother looks up at the stars to map constellations, I prefer looking at the ticker tape to find patterns.
I am Dhruv Saxena. I am a debater, a coder, and an athlete. I’m not just here to watch the economy evolve; I’m learning to write the code that will drive it.
View My Work
"The important thing is not to stop questioning."
Surrounded early on by discussions of market shifts and global finance, I quickly realized that understanding business terminology wasn't enough - I wanted to build the mathematical and computational engines behind it.
Knowing the words isn't the same as writing the code.
That realization pushed me to stop just watching the markets and start learning how to build the tools that analyze them. I moved away from relying on exposure and began actively studying Computer Science. Currently, I am focused on learning Python, building scripts that track stock volatility, and translating market chaos into algorithmic logic.
I have learned that the discipline required to master a complex chord progression on my guitar is the same discipline needed to debug an algorithm. The strategy required to predict an opponent’s bowling in cricket is the same mindset needed to spot a market trend.
Class X: 94%
Class XII: 97% (Projected)
SAT: 1440 95th percentile
Math 780 98th percentile
EBRW 660 86th percentile
TOEFL iBT: 5.5/6
Focus: STEM & Finance
Download Resume
Problem: Novice investors often mistake high-volatility "hype" stocks for safe, stable investments, leading to capital loss.
Action: Used Python to calculate 30-day historical price variance, identifying underlying volatility often missed by the market.
Result: Developed a data-backed risk assessment framework to objectively evaluate asset stability.
Problem: Purely statistical trading models often fail because they cannot account for the sudden, unpredictable impact of breaking news and geopolitical shifts.
Action: Built an analytical model tracking Beta to measure specific stock volatility against broader market movements.
Current Focus (V2.0): Pure statistics can't predict human panic. Building a Python-based AI agent that evaluates historical stock trends (variance, covariance) while simultaneously running sentiment analysis on financial headlines, to test whether combining the two gives a sharper read on risk than either alone.
Problem: "Buy Now, Pay Later" apps often use deceptive "Dark Pattern" UI designs to trap students in impulse debt cycles.
Action: Audited the checkout flows of 3 Indian BNPL apps and surveyed 52 Gen Z respondents (ages 16–21) to identify specific predatory design elements and measure their real-world effect.
Result: 44% of respondents were regular BNPL users (3+ times a month). Of those exposed to false-urgency prompts, 82% reported making an impulse purchase; of those hit by hidden late fees, 65% went on to miss a payment. The paper proposes concrete fixes - a mandatory cooling-off window before checkout, a ban on pre-selected auto-pay, a clear "cost-of-credit" label, and income-verified age-gating.
Agentic AI and Data Engineering Intern
The Context: To understand what happens before an AI model is actually trained. I joined the engineering team to understand the raw data foundation that powers supervised learning.
The Work: Supported the core machine learning pipeline by manually segregating, labeling, and quality-checking training data - roughly 30-40 records a day.
The Takeaway: Caught a data point that looked like an error - a 50% overnight price drop and traced it back to an ordinary stock split rather than a data quality issue. It taught me that good data work isn't just about spotting anomalies; it's about knowing which ones are actually meaningful and which are just the market doing something normal.
Quality Assurance (QA) Intern
The Context: Working inside a CMMI Level 3-certified engineering environment gave me a front-row seat to the actual software development lifecycle (SDLC) and how products go from code to the end-user.
The Work: Responsible for manual "smoke testing". Before the senior developers did their final checks, I tested new web and mobile apps exactly as an everyday user would, actively trying to find broken links, navigation errors, and layout bugs.
The Takeaway: Learned that writing the code is only half the battle. I learned how to clearly document and explain the exact bugs I found so the developers could easily fix them before the app went live to real users.
Fundraising & Organizing Lead,
in partnership with the NGO
Active Volunteer,
Hyderabad Chapter
Delegate & Mentor
Open to conversations on quantitative finance, behavioral economics, and collaborative research.
Send an Email
