Shivesh Prakash

Research

1. 

   From Word Counts to Context: Topic Models for Asset Pricing

   Kevin Foley, Jonathan Hartadi, Shivesh Prakash, Swapnil Vatsal; advised by Ali Kakhbod

   Working paper, 2026. Authors in alphabetical order; I’m the corresponding author.

   Does reading financial news in context build better risk factors? We compare LDA’s word counts with a sentence transformer on 394K articles, holding a Sparse IPCA pricing pipeline fixed. The point-in-time follow-up, “Reading the News Without Hindsight” (forthcoming), contrastively fine-tunes NoLBERT on 1.26M articles, lifting top-1 matching on unseen 2022–24 news from 3% to 86%. Its price + news ensemble earns a net Sharpe ratio of 1.19 on held-out 2022–24 data.

   • Project page
   • arXiv
   • SSRN
   • Code
2. 

   FOMO: FLEX Options and Market Outcomes

   Shivesh Prakash, Frank Ferstler, Nicholas Pezolano

   Working paper, 2026. UC Berkeley MFE industry project with NewMark Risk.

   FLEX options let institutions choose their own strikes and expiries. On a 910-name panel, matched difference-in-differences shows that FLEX events raise listed-option open interest by 0.306 log points (about 36%) and cluster in high-implied-volatility names. We also stress-test volatility strategies around these events for trading costs.

   • Project page
   • Paper
   • Technical report
3. 

   In progress

   Zero-shot FX forecasting with time-series foundation models

   Millennium Global, FX Execution Team (MillTech). UC Berkeley MFE industry project, Oct–Dec 2026.

   Testing whether zero-shot forecasts from TiRex-2, an xLSTM foundation model given option-implied covariates, beat classic FX momentum. The backtests are point-in-time and walk-forward on G10 and a 20-pair universe that includes emerging markets; forecast quantiles map to positions, and interval calibration is checked through crisis periods.

4. 

   A User-Tunable Machine Learning Framework for Step-Wise Synthesis Planning

   Shivesh Prakash, Nandan Patel, Hans-Arno Jacobsen, Viki Kumar Prasad

   Digital Discovery, 2026. First author.

   MHNpath plans chemical syntheses one step at a time, ranking reaction templates with modern Hopfield networks. Chemists can weight routes by cost, temperature and toxicity, and it solves 85.4% of PaRoutes targets.

   • Paper
   • arXiv
   • Code
5. 

   VIXtrader: VIX Forecasting and Futures Trading

   Shivesh Prakash, advised by Ing-Haw Cheng

   Working paper, 2025.

   An LSTM on macro and ARMA features forecasts the 34-day VIX, cutting MAE by 10% and doubling R² against the baseline. Dynamic-leverage strategies built on the forecast reach backtested Sharpe ratios 2.6 times those of passive benchmarks.

   • Paper
6. 

   In progress

   Proving a bond-futures hedge correct in Lean

   Citi. UC Berkeley MFE industry project, Oct–Dec 2026.

   Writing a machine-checked proof, in the Lean theorem prover, that a bond-futures hedge is correct, then extending it to regime switches, jumps and options.

7. 

   NanoTick: a limit-order-book replay engine

   Independent project. C++20, with Parquet, PyArrow and Numba tooling.

   Normalizes raw Nasdaq TotalView-ITCH data into Parquet and replays the displayed L3 order book in C++20: 30M BX messages at 4.8M messages per second on an M1 Mac, with an average p99 of 284 ns per book operation and no order-state errors across a full trading day.

   • Code
8. 

   SEE-2-SOUND: Zero-Shot Spatial Environment-to-Spatial Sound

   Rishit Dagli, Shivesh Prakash, Robert Wu, Houman Khosravani

   SIGGRAPH 2025 Posters; ICML 2024 workshop.

   Generates 5.1 surround sound for images and videos with no training: find the regions that make sound, place them in 3D, generate audio for each, and mix them into a spatial track.

   • Project page
   • arXiv
   • Code
9. 

   Equivariant Graph Neural Network for Rapid Transition State Structure Predictions in Chemistry

   Shivesh Prakash, Hans-Arno Jacobsen, Viki Kumar Prasad

   Working paper, 2025. Best Poster at the UofT Data Sciences Institute’s SUDS program.

   A LeftNet-based equivariant graph neural network predicts transition-state geometries from reactants and products, with a mean absolute error of 18.6 pm.

   • Paper
10. 

    Efficient Training of Transformers for Molecule Property Prediction on Small-scale Datasets

    Shivesh Prakash

    arXiv, 2024. Sole author.

    A GPS Transformer with standard self-attention predicts whether drugs cross the blood–brain barrier from a small dataset, reaching 78.8% ROC-AUC on BBBP, 5.5% above the previous state of the art.

    • arXiv

• Beyond the Visible. Diffusion-based denoising of hyperspectral images for the FINCH CubeSat, with the University of Toronto Aerospace Team. arXiv, 2024. arXiv
• Material sensing. Sub-second hyperspectral capture and a DNG-to-RGB demosaicing pipeline, with trained material classifiers. Code
• DygnosTech. GAN-based prediction of drug side effects with WebAssembly-optimized models; won MetroHacks. Code

Experience

1. Oct–Dec 2026

   Millennium Global

   Quant Researcher (part-time), FX Execution Team (MillTech)

   UC Berkeley MFE industry project. Testing zero-shot foundation-model forecasts against classic FX momentum on G10 and emerging-market currencies.

2. Jun–Aug 2026New York

   JPMorganChase

   Quantitative Research Summer Associate, Quant Trading & Research

   • Cut T+1 forecast error (MAE) by 15% against a seasonal baseline, using a 50K-per-day transaction stream and microstructure features.
   • Built a real-time forecasting engine with settlement prediction, clustering and path inference from flows.
   • Mapped unstructured news into directional NLP factors to explain event-driven regime switches in capital flows.

3. Sep 2025 – Feb 2026Mumbai

   EY

   Intern, Treasury Quant Team

   Built a privacy-first retrieval-augmented query engine for corporate treasury that helped win four new clients.

4. Jun–Sep 2025Delhi

   Pace Stock Broking Services

   Quantitative Research Intern, HFT Desk

   Clustered co-moving stocks from minute-level returns (PCA/LSTM + HDBSCAN); cluster returns showed 3.3% autocorrelation.

5. Jun 2024 – Aug 2025Toronto

   University of Toronto

   Research Assistant

   • Processed 9 TB of WSJ and NYT news to study the excess bond premium (draft), and modeled short-put volatility surfaces (paper).
   • Used symbolic differentiation to solve a large PDE system in a bank-run model (notes).
   • Designed the equivariant graph neural network for transition-state prediction listed under Research.

Competitions and honors

• JEE All India Rank 389, top 0.04% of candidates
• Physics Olympiad, first in state and national top 1%
• Regional Math Olympiad qualifier, twice
• International Scholar, University of Toronto ($100K)
• Best Poster, UofT Data Sciences Institute SUDS
• Passed Level I of the CFA Program

Education

1. Expected Mar 2027

   University of California, Berkeley

   Master of Financial Engineering, Haas School of Business. GPA 3.8/4.0.

   Stochastic Calculus with Asset Pricing, Derivatives, Fixed Income, High-Frequency Finance.

2. May 2025

   University of Toronto

   Honours BSc in Computer Science, minor in Statistics. GPA 3.88/4.0.

   Machine Learning (A+), Math of Finance (A+), Graduate Optimization (A), Probabilistic Learning (A). DSI Award; three-time Dean’s List.