Skip to content
View jeragilo's full-sized avatar
🎯
Focused
🎯
Focused

Block or report jeragilo

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
jeragilo/README.md

Jesús Gil

Computer Scientist & Applied Mathematician
Quantum Computing · Machine Learning · High-Performance Computing

I build research-driven computational systems at the intersection of quantum computing, machine learning, and scientific computing. My work focuses on reliability under real constraints: quantum noise, limited circuit depth, computational scaling, model stability, and hardware-aware performance.

Research Publications & Manuscripts

Beyond Parity: Measurement Representation and Decoder Capacity in Noise-Robust Hybrid Quantum Neural Networks

Submitted to Quantum Machine Intelligence (Springer Nature)
Coauthored with Dr. David Hart, East Carolina University.

Experimental research on quantum measurement representations, classical decoder capacity, and noise robustness in hybrid quantum neural networks, including controlled benchmarking and IBM Marrakesh hardware validation.

From Quantum Possibility to Quantum Utility: A Cross-Domain Systematic Review and Six-Level Evidence Framework

Submitted to IEEE Access
Coauthored with Dr. Nasseh Tabrizi, East Carolina University.

A systematic review synthesizing 349 retained reviews across eight quantum computing domains and introducing a six-level evidence framework for evaluating practical quantum utility.

Featured Work

Noise-Robust Hybrid Quantum Neural Networks

Master’s thesis research — East Carolina University
noise-robust-hqnn-framework

A reusable experimental framework for evaluating and improving hybrid quantum–classical neural networks under NISQ-era noise.

  • Built with Qiskit, Qiskit Aer, Cirq, PennyLane, scikit-learn, NumPy, and Matplotlib
  • Implements depolarizing, bit-flip, phase-flip, and amplitude-damping noise models
  • Evaluates learned classical readout, multi-observable quantum features, architecture search, noise-aware objectives, repeated trials, and statistical validation
  • Improved evaluated noisy classification performance from approximately 0.37 with fixed parity readout to approximately 0.82–0.85 with learned and multi-observable readouts
  • Explores how measurement design, entanglement structure, and classical post-processing affect robustness in hybrid quantum learning systems

Graduate Quantum Computing Curriculum & QREP Framework

Ongoing curriculum and research-software development — East Carolina University

Collaborating with university faculty on a 15-week graduate quantum computing curriculum and a 12-laboratory instructional framework connecting quantum theory, Qiskit programming, simulation, and IBM Quantum workflows.

  • Developing Jupyter notebooks, lecture materials, laboratory guides, and technical demonstrations.
  • Engineering a reusable Python package for quantum circuit experiments, statistical analysis, benchmarking, and reproducible results.
  • Building automated tests, backend inspection, transpilation checks, resource-budget planning, and guarded hardware-submission workflows.
  • Development status: Course materials and laboratories remain under development and faculty review. Hardware planning and offline validation should not be interpreted as completed physical QPU execution for every lab.

CUDA Particle Simulation

cuda-particle-simulation-hpc

GPU-accelerated particle simulation developed and benchmarked on the Bridges2 supercomputer.

  • CUDA C/C++ implementation with serial CPU baseline
  • GPU parallelization and memory-access analysis
  • SLURM execution and performance benchmarking
  • Correctness validation before performance measurement

Parallel Scientific Computing with MPI

High-Performance-Parallel-Pi-Approximation-Using-MPI-and-SLURM

Distributed numerical computation using C++, MPI, and SLURM on Bridges2.

  • Benchmarked scaling across 1, 12, 24, and 48 cores
  • Achieved approximately 10× speedup at 12 cores
  • Analyzed efficiency loss and communication overhead as parallelism increased

LSTM Forecasting & ML Deployment

lstm-stock-forecasting-system

End-to-end machine-learning system combining PyTorch, ONNX, Flask, React, and financial time-series features.

  • LSTM training and evaluation pipeline
  • ONNX export and inference validation
  • Flask API and React frontend
  • Technical indicators including RSI, SMA, EMA, momentum, volatility, and MACD

Research Portfolio Platform

AI-Research-Portfolio-System

A deployed Next.js + TypeScript research portfolio with dynamic routing, reusable components, and Vercel deployment.

Blockchain Identity Prototype

Blockchain-identity-verification

Experimental decentralized identity-registration system using Solidity, Hardhat, Ethereum/Sepolia, ethers.js, Next.js, TypeScript, and MetaMask.

Technical Focus

Quantum & Research
Qiskit · Cirq · PennyLane · NISQ noise modeling · Hybrid quantum–classical algorithms · Quantum machine learning

Machine Learning & Scientific Computing
Python · PyTorch · scikit-learn · NumPy · Pandas · ONNX · Jupyter

High-Performance Computing
C++ · CUDA · MPI · SLURM · GPU computing · performance benchmarking

Software & Platforms
Git · Linux · SQL · Flask · React · Next.js · TypeScript · Docker · AWS · Azure

Education

  • M.S. Computer Science — East Carolina University
  • B.S. Mathematics — North Carolina Wesleyan University

Links

LinkedIn · GitHub

Pinned Loading

  1. noise-robust-hqnn-framework noise-robust-hqnn-framework Public

    Master’s thesis research framework for noise-robust hybrid quantum–classical neural networks (HQNNs), evaluating reliability, architecture design, and deployment strategies on NISQ hardware.

    Python 1

  2. lstm-stock-forecasting-system lstm-stock-forecasting-system Public

    A full-stack time-series forecasting system for equity markets using LSTM neural networks, ONNX inference, and quantitative indicators.

    Python 1

  3. AI-Research-Portfolio-System AI-Research-Portfolio-System Public

    A production-ready AI research portfolio platform demonstrating modern frontend engineering, cloud deployment, and research presentation workflows.

    TypeScript 1

  4. Blockchain-identity-verification Blockchain-identity-verification Public

    Decentralized identity verification dApp built with Solidity smart contracts and a Next.js frontend. Enables users to securely register identity hashes on Ethereum via MetaMask, demonstrating full-…

    JavaScript 1

  5. cuda-particle-simulation-hpc cuda-particle-simulation-hpc Public

    High-performance particle simulation implemented with CUDA, including CPU vs GPU benchmarking and large-scale execution on the Bridges2 supercomputer. Focuses on parallel algorithms, performance op…

    Cuda 1

  6. High-Performance-Parallel-Pi-Approximation-Using-MPI-and-SLURM High-Performance-Parallel-Pi-Approximation-Using-MPI-and-SLURM Public

    Scalable parallel π approximation implemented with MPI and executed on an HPC cluster using SLURM. The project explores distributed Monte Carlo methods, communication overhead, and strong scaling b…

    C++ 1