Skip to main content
Daksh UserDU
Open to opportunities

Daksh User

@dakshuser3

Undergraduate researcher at IIT Roorkee studying information leakage and safety trade-offs in multi-agent AI.

India
Message

What I'm looking for

I’m interested in the mathematical foundations of multi-agent AI, especially how agents communicate, make decisions, and coordinate, and what can be established from observable behavior. I’m drawn to work on covert communication, monitoring limits, and safety–performance trade-offs.

At IIT Roorkee, in collaboration with COSIC, KU Leuven, I’m researching information flow, observational limits, and safety–performance trade-offs in LLM-based multi-agent systems. My work includes investigating how agents might retrieve sensitive information and coordinate communication, and comparing interaction protocols.

At COSIC, KU Leuven, I developed finite-model results on hidden information flow under passive monitoring and designed controlled sender–receiver experiments. I’ve also worked on post-quantum cryptography, including a memory-conscious ARM Cortex-M4 implementation of FAEST with Bocconi University, and write about mathematics, epistemology, logic, and intelligent systems.

Experience

Work history, roles, and key accomplishments

IIT Roorkee logoIR
Current

Undergraduate Thesis Research

Jan 2026 - Present (10 months)

Conducting research on mathematical foundations of LLM-based multi-agent systems, focusing on information flow, observational limitations, strategic coordination, and safety-performance trade-offs.

Multi Agent Systems
Information Theory
Decision Theory
Game TheoryLLM Security
Safety Evaluation
COSIC, KU Leuven logoCL

Research Intern

COSIC, KU Leuven

Jun 2026 - Aug 2026 (2 months)

Independently explored LLM multi-agent security and formulated an information-theoretic investigation of passive monitoring limitations. Developed finite-model results and designed controlled LLM sender-receiver experiments.

LLM Security
Information Theory
ProbabilityExperimental Design
Data Analysis
Bocconi University logoBU

Undergraduate Researcher

Bocconi University

Jan 2026 - Apr 2026 (3 months)

Worked on a memory-conscious ARM Cortex-M4 implementation of the FAEST post-quantum signature scheme, connecting cryptographic specifications to embedded memory and performance constraints.

Education

Degrees, certifications, and relevant coursework

Indian Institute of Technology Roorkee logoIR

Indian Institute of Technology Roorkee

B.Tech., Engineering Physics

2023 -

Activities and societies: Mathematics: Real Analysis I–II, Linear Algebra, Number Theory, Graph Theory, Probability & Statistics, Complex Analysis, ODEs/PDEs, Fourier and Laplace methods, Mathematical Physics. Mathematical physics: Quantum Mechanics, Quantum Theory of Solids, Statistical Mechanics, General Relativity, Numerical Physics. Independent study: Abstract Algebra, Information Theory, Category Theory, Functional Analysis, Topology.

Pursuing a B.Tech. in Engineering Physics with a minor in Mathematics and Computing, covering advanced mathematics and mathematical physics.

Tech stack

Software and tools used professionally

Availability

Open to opportunities

Location

India

Authorized to work in

Salary expectations

140k+ INR

Job categories

Skills

Multi Agent SystemsAI Safety
Information Theory
Decision Theory
Game TheoryProbability
Stochastic Process
Cryptanalysis
Post Quantum Cryptography
Side Channel Resistance
Homomorphic Encryption
LLM Evaluation
Information Flows
Mathematical ModelingC++

Interested in hiring Daksh?

You can contact Daksh and 600K+ other talented remote workers on Himalayas.

Message Daksh

People also viewed

View all talent

Get matched with your dream remote job

Sign up now and join over 1,000,000+ remote workers who receive personalized job alerts, curated job matches, and more for free!

Sign up
Himalayas profile for an example user named Frankie Sullivan