
Dinis Felgueiras
@dinisfelgueiras
ML researcher at INESC TEC building LLM-guided optimization for photovoltaic energy forecasting and battery dispatch.
What I'm looking for
At INESC TEC, I built a Python pipeline that connects LightGBM photovoltaic energy forecasting with battery dispatch optimization. I also developed an LLM-guided parameter tuning system that uses reward-based feedback.
At Instituto Superior Técnico, my MSc thesis compares neural architectures with a learnable Pretty Good Measurement readout for pattern recognition. I developed differentiable density matrix embeddings and relaxed measurement operators in PyTorch, and evaluated the models on MNIST and FashionMNIST under reduced-data and class-imbalanced conditions.
As a Research Intern at Instituto de Telecomunicações, I analysed distributed acoustic sensing data from submarine optical fibre systems for seismic event classification. I used Python, MATLAB, NoisePy, and ObsPy to examine the data and signal correlations.
For a research project at Instituto Superior Técnico, I designed distributed circuit decompositions for Shor’s algorithm and implemented distributed quantum simulations using Qiskit. I’m seeking roles in machine learning engineering, data science, or AI.
Experience
Work history, roles, and key accomplishments
ML Researcher
Instituto Superior Técnico
Jan 2024 - Present (2 years 9 months)
Implemented and compared four neural architectures (MLP, CNN, KAN, QCNN) with a learnable Pretty Good Measurement (PGM) readout layer replacing softmax, targeting low-data and class-imbalanced regimes. Investigated PGM-based readout layers across different neural architectures on MNIST and FashionMNIST classification benchmarks.
ML Researcher
INESC TEC
Jan 2024 - Present (2 years 9 months)
Built an end-to-end Python pipeline integrating a LightGBM recursive forecasting model with battery dispatch optimization for photovoltaic energy production. Developed an LLM-guided parameter tuning system using reward-based feedback to adapt PV forecast correction parameters.
Research Intern
Instituto de Telecomunicações
Jan 2023 - Dec 2024 (1 year 11 months)
Analysed distributed acoustic sensing (DAS) data from submarine optical fibre systems for seismic event classification. Investigated correlations between optical signal propagation parameters and seismic signatures using Python and MATLAB.
Research Project
Instituto Superior Técnico
Jan 2023 - Dec 2023 (11 months)
Designed distributed circuit decompositions to reduce qubit requirements in Shor's algorithm, analysing gate-complexity scaling and communication–computation trade-offs. Implemented distributed quantum simulations using Qiskit.
Education
Degrees, certifications, and relevant coursework
Instituto Superior Técnico
MSc, Electrical and Computer Engineering
2024 -
Grade: 16.5/20
Pursuing an MSc in Electrical and Computer Engineering with a GPA of 16.5/20. Relevant coursework includes Optimization and Algorithms, Quantum Information Processing, Deep Learning, and Machine Learning.
Instituto Superior Técnico
BSc, Applied Mathematics and Computation
2021 - 2024
Completed a BSc in Applied Mathematics and Computation with coursework in Computability and Complexity, Mathematical Logic, Discrete Mathematics, and Programming.
Availability
Location
Authorized to work in
Salary expectations
Job categories
Interested in hiring Dinis?
You can contact Dinis and 90k+ other talented remote workers on Himalayas.
Message DinisGet matched with your dream remote job
Sign up now and join over 250,000+ remote workers who receive personalized job alerts, curated job matches, and more for free!
