I'm Pradyot, a Computer Engineering and Mathematics (Program II) undergraduate at Indiana University Bloomington, expecting to graduate in May 2028. My interests are quantitative finance, machine learning, and scientific computing. I build asset-pricing experiments, study numerical error in quantum simulations, and work on signal analysis and modeling for optometry research with Prof. S. P. Srinivas and Prof. Yuxi Hong. Across those projects, the question is the same: what does the experiment actually establish? My paper on training-data leakage in LLM-assisted portfolio optimization was accepted to QNLP-AI 2026 for an oral presentation. Based in Bloomington, IN.
The research journey
How the reading turned into projects, and the projects into a direction. Every project names what it was built on.
- 1Read / learned2026
Linear algebra across the projects
Coursework and reference reading connect eigenstructure, projections, and SVD to PCA, factor models, and quantum gates. The same tools keep reappearing in different applications.
- 2Read / learnedSpring 2026
High Performance Computing (ENGR-E 517, graduate)
Memory hierarchy, parallelism, and the habit of measuring instead of guessing. The course that made me ask what the cache is actually doing.
- 3
- 4Read / learned2026
Empirical asset pricing: Cochrane, Fama-French, Baba-Yara
Discount Rates (2011), Common Risk Factors (1993), and In Search of Sparsity. The factor-zoo problem and why machine learning has to be disciplined by economics.
- 5Read / learned2026
MITx Foundations of Modern Finance + Ruppert
CAPM, APT, and the statistics of financial data — MITx 15.415.1x completed with honors — paired with Ruppert's Statistics and Data Analysis for Financial Engineering for the empirical machinery.
- 6Read / learned2026
ARPM Bootcamp (onsite)
A week of buy-side quantitative methods taken onsite — the ARPM 'Checklist' end to end: estimation, projection, pricing, aggregation, and portfolio/risk construction. The bridge from textbook asset pricing to how it is actually practiced.
- 7
- 8Read / learned2026
Quantum computing: Sutor, Wong, Nielsen-Chuang
Sutor and Wong for the foundations, with Nielsen-Chuang as a reference: state vectors, unitary gates, and measurement. Conditional normalization is nonlinear; the outcome-averaged density-matrix map remains linear.
- 9
- 10
- 11Shipped2026
Paper: Training Data Set Effects in LLM-Assisted Quantum Portfolio Optimization
Accepted to QNLP-AI 2026 for an oral presentation. Shows that an LLM's apparent portfolio alpha collapses once the model is genuinely out-of-sample — the edge was training-horizon look-ahead, not signal.
- 12
- 13
- 14
- 15Read / learnednext
Toward a PhD in quantitative finance
A research arc where numerical rigor, modeling depth, and systems thinking all have to coexist — and every claim still traces back to a run ID.
Recent writing
- Building an ethics tool inside the tab people already use
AI Ethics Coach is a Chrome MV3 extension that reviews selected text and returns practical flags without pretending to replace policy or judgment.
- Captain Whiskers: a trading agent that optimizes portfolios with a quantum circuit and signs them with post-quantum crypto
A hackathon build that had no business working: an autonomous on-chain trading agent fusing variational-quantum portfolio optimization, on-chain identity, a risk router, Gemini-driven signals, post-quantum signatures, and an eleven-node Byzantine-fault-tolerant verification layer. Here is the architecture, the math, and which parts were real versus theater.
- Auditing a multi-terabyte filesystem without melting the laptop that audits it
disk-archival-toolkit streams multi-million-row filesystem inventories, classifies storage tiers, and emits budget-aware archival manifests, all in bounded memory. A short note on why the streaming constraint is the whole design, not a detail.
- Our first forecasting model lost to zero. We kept it.
A Data Science Club learning project: lagged international returns, chronological validation, a negative baseline result, and a final period we have not touched.
Project gallery
Selected projects, with implemented work distinguished from ongoing research.
Optometry research: blink signals and ocular modeling
Developing offline EOG blink-segmentation workflows, reviewing event timing, and preparing reproducible ocular-modeling experiments. The emphasis is on participant-separated evaluation and reviewable outputs, not a clinical-performance claim. Research notes.
qhpc_cache — cache behavior of quantitative-finance kernels
An empirical L1-cache characterization of four numerical-finance kernels (Cholesky, Monte Carlo, GARCH, GEMM) on AMD EPYC, instrumented with PAPI hardware counters on BigRed200. Every claim traces back to a counter measurement, not a vibe.
RegimeFactorZoo — machine-learning factor models with a regime twist
Reconstructed SMB and HML from CRSP/Compustat, then reproduced Bryzgalova, Huang and Julliard's Bayesian factor-selection output to approximately machine precision under the authors' replication setup. The volatility-regime extension remains preliminary until its matched-size null comparison is run. The repository is public; licensed-data experiments still require the appropriate data access.
Quantum computing in the Hong Lab
From gate algebra to a device-calibrated simulation study: comparing readout correction and zero-noise extrapolation while separating circuit error from Hamiltonian-discretization error. These are simulations, not a claim of hardware quantum advantage. Methods and results.
Captain Whiskers — autonomous on-chain trading agent
A hackathon build: an autonomous trading agent with variational-quantum portfolio optimization, on-chain identity, risk limits, and an eleven-node Byzantine-fault-tolerant verification layer. Yes, the cat has a risk router.
legal-document-intelligence — agentic diligence over messy documents
A retrieval-augmented pipeline (LangGraph + Qdrant, served on vLLM) for property and M&A document review, tuned for AMD MI300X. Built for the unglamorous reality that most "AI" work is making retrieval not lie to you.
About me
I started in visual systems at SCAD, moved into engineering at IU, and added a Mathematics major because I wanted the proof-heavy foundation, not just the applied shortcuts. That combination now points at one goal: a PhD in quantitative finance, working on methods that are measurable, reproducible, and useful outside a backtest.
This fall, that means moving between optometry signal-analysis experiments, numerical-methods work, and empirical finance. As Head of Technology for IU's Data Science Club, I also help turn research questions into small projects with baselines, tests, and code review. My earlier design and startup work still shapes how I explain and build things; the personal section keeps that side visible.