PhD applicant for Fall 2027

Reinforcement learning for quantum information systems.

I study how learned policies and algorithmic methods can solve structured decision problems in quantum networks and quantum computing—from entanglement routing and decoding to circuit design for chemistry and materials.

Download CV Email me

Research trajectory

A connected research progression

Select a node to see how each stage builds on the previous one.

About

I'm a graduate student at Carnegie Mellon University's Information Networking Institute, finishing an MS in Information Technology – Information Security in May 2027. My undergraduate degree is in Computer Engineering, where I graduated valedictorian with a 3.83/4.00 GPA.

My path into quantum computing came through networking and security — I got interested in quantum networks as a distributed-systems problem first, and found that the reinforcement-learning tools I already knew transferred more directly than I expected. That's shaped how I think about the field since: I'm less interested in quantum computing as a topic to study from the outside, and more interested in it as a set of concrete, structured decision problems — routing, decoding, circuit construction — that a learned policy can plausibly do better than a hand-derived heuristic, if the problem is posed carefully.

I'm currently applying to PhD programs for Fall 2027, looking for a research home where that approach — RL and algorithmic thinking applied to real quantum-hardware problems — has room to grow, whether that ends up being quantum error correction, quantum chemistry algorithms, or something that blends the two.

Tanish Singh Rajpal standing on a palm-lined campus walkway during a visit to Stanford University Campus visit
At Stanford University during a campus visit.

Research

Reinforcement-learning-based routing for heterogeneous quantum networks In progress

May–October 2026

Independent Study, Information Networking Institute, CMU — Advisor: Prof. Michael Mattarock

Quantum networks move entanglement, not bits, and the thing you're optimizing for — fidelity — degrades with every hop and every noisy component in a way that classical shortest-path routing was never built to reason about. This project asks a narrow, concrete version of a bigger question: can a decentralized learned policy, using only local information, route entanglement through a noisy quantum network as well as or better than a centralized heuristic that can see the whole topology?

I built a router using Proximal Policy Optimization, implemented in NetSquid, where each node runs its own agent making decisions from a fixed, scale-invariant observation and action space — meaning the same trained policy generalizes across network sizes without retraining. I benchmarked it against two centralized baselines: a negative-log-fidelity-weighted Dijkstra's algorithm, and simple hop-count routing, across both homogeneous and heterogeneous quantum hardware models.

≥99%mean fidelity, homogeneous grids 3×3–10×10
90–93%mean fidelity under heterogeneous hardware noise
5 / 6tested configurations at 100% routing success
3independent training seeds per result

The one configuration that didn't hit 100% success is worth stating plainly rather than smoothing over: the 10×10 homogeneous grid reached an 84.4% success rate with a wide ±27.0% spread across seeds — a real signal about where decentralized, locally-observed policies start to strain as network size grows, and a concrete direction for follow-up work rather than a clean result to round up.

Research preprint and journal submission targeted for October 2026. Simulation code, trained models, and results will be released as an open-source repository on completion.

A hybrid architecture for post-quantum EAP-TLS In progress

August 2026–April 2027

Master's Thesis, Information Networking Institute, CMU — Advisors: Prof. Pedro Bustamante & Prof. Michael Mattarock · Reader: Prof. Patrick Tague

Post-quantum cryptographic certificates are large enough to break assumptions that classical network protocols quietly depend on. This thesis looks at one specific failure mode: oversized ML-DSA (post-quantum) certificates used in EAP-TLS authentication can trigger silent Path MTU Discovery blackholes — connections that fail without any clear error, because a packet is dropped somewhere and nothing reports it.

The proposed fix combines an edge QUIC-terminating proxy at the Wireless LAN Controller with a Quantum Key Distribution–secured wired backbone, inside a hybrid WPA-Enterprise 802.1X architecture. The thesis is organized around four testable hypotheses, simulated in NS-3:

  • Handshake latency — targeting a ≥40% reduction in Flow Completion Time under elevated wireless error rates.
  • RADIUS server compute offload under the proxy architecture.
  • IPv4/IPv6 baseline isolation, to separate protocol-specific effects from the core result.
  • Sustained QKD key-delivery capacity under high-density authentication bursts.

Thesis defense and final report targeted for Spring 2027. Simulation code and datasets will be released via a CMU-hosted GitHub repository and the CMU KiltHub data repository.

Publications

Journal articles

Rajpal, T. S., & Naithani, A. (2026). NeuroCrypt: A Neuro-Symbolic AI Ecosystem for Advanced Cryptographic Data Security and Transmission. IEEE Transactions on Artificial Intelligence, 7(1), 512–521.

NeuroCrypt pairs a rule-based cryptographic engine with a deep neural network to dynamically generate symmetric keys, aiming to make encryption harder to model and attack than a fixed algorithm. Evaluated against classical ciphers (AES, DES) and neural-network-based attacks, it showed a 25% increase in measured security resilience and a 15% reduction in key-management overhead.

doi.org/10.1109/TAI.2025.3577605

Conference papers

Naithani, A., & Rajpal, T. S. (2025). Blockchain-Based Digital Health Passport: A Decentralized Framework for Secure and Interoperable Health Data Management. 2025 IEEE Zooming Innovation in Consumer Technologies Conference (ZINC).

A decentralized health-record system built to guarantee data immutability and patient ownership through a unique identifier and OTP-based consent mechanism, designed for secure interoperability across providers at national scale. It uses a hybrid on/off-chain storage model with public-key encryption and role-based access control, aiming to keep large medical records both secure and compliant without sacrificing scalability.

doi.org/10.1109/ZINC65316.2025.11100281

Naithani, A., Rajpal, T. S., & Mistry, V. (2024). TAV-Cart: Transaction Assistance and Validation Cart System for Effortless Shopping and Checkout. 2024 International Conference on Advances in Computing, Communication and Materials (ICACCM).

A system design for a smart shopping cart intended as a lower-cost alternative to RFID-based carts: dual barcode scanners and a strain-gauge weight-validation system cross-check that what's scanned matches what's physically in the cart, an LED strip gives real-time feedback, and checkout happens directly through UPI — India's national instant-payment system — removing the need for a cashier line. This is a feasibility and system-architecture study, not a deployed retail pilot.

doi.org/10.1109/ICACCM61117.2024.11059075

Naithani, A., Jain, V., Mistry, V., & Rajpal, T. S. (2024). AI Based Goalkeeper for Penalty Shot Prediction Using RNN and Pattern Matching Algorithm. 2024 12th International Symposium on Digital Forensics and Security (ISDFS).

Built a predictive model — a recurrent neural network combined with a pattern-matching module — that learns a penalty taker's shot-direction tendencies from their history and predicts where they'll shoot next. Tested against four shooting strategies (a fixed repeating pattern, always the same side, fully random, and adaptively shooting away from the keeper's last dive) at 30 shots each: fully random shooting produced the lowest save rate of any strategy, directly showing that unpredictability is a penalty taker's strongest defense against a model trying to read them. Validated against Cristiano Ronaldo's recorded penalty history from the 2014–15 through 2017–18 seasons: the AI model saved 17 of 53 penalties in situations where the real goalkeepers on record saved 13 — a modest but real improvement, and an initial prototype rather than a production system.

doi.org/10.1109/ISDFS60797.2024.10527307 Code repository

Selected projects

Quantum portfolio optimization using QAOA — Team Lead

Yale University's YQuantum 2026 Hackathon — 3rd place
Markowitz portfolio model
QUBO formulation
Ising Hamiltonian
QAOA circuit

Built a quantum computing approach to optimizing insurance investment portfolios: mapping classical Markowitz portfolio theory into a QUBO formulation, then converting that into an Ising Hamiltonian solvable with the Quantum Approximate Optimization Algorithm. Led the team's circuit construction and problem formulation work over the hackathon.

View repository
April 2026

Experience

Vulnerability Assessment and Penetration Testing Intern

Newton's Apple, Pune, India

Ran penetration testing on a confidential oil company's Android application using Burp Suite and Metasploit, identifying and helping remediate three critical vulnerabilities that protected sensitive operational data. Static Java code analysis work, done in collaboration with the development team, contributed to a 40% reduction in common security flaws in subsequent releases.

October 2023–January 2024

Recognition and leadership

Student Ambassador

Information Networking Institute, Carnegie Mellon University
October 2025–present

IBM Qiskit Global Summer School 2026 · Excellence Badge

Completed Labs 0–4 and the Advanced Track.
2026

Third place · Yale YQuantum Hackathon

Quantum portfolio optimization project
2026

First place · Crypto Quest Cybersecurity Competition

Tech-Fest Ambiora
2024

Contact

Let's talk about quantum information systems.

I am applying to PhD programs for Fall 2027 and welcome conversations about quantum networking, reinforcement-learning-based error correction, and quantum algorithms.

trajpal@andrew.cmu.edu

GitHub LinkedIn CV