Ph.D. Candidate · CBPF · Rio de Janeiro

Phelipe Darc

ASTROPHYSICS / MACHINE LEARNING / MULTI-MESSENGER ASTRONOMY

I am an astrophysics researcher focused on the electromagnetic counterparts of compact-object mergers, including binary black holes embedded in AGN disks and kilonovae from neutron star–black hole and neutron star–neutron star mergers. My work combines theoretical modeling, machine learning, and simulation-based inference to understand the most energetic transients in the Universe. I also develop physics-informed open-source tools for the multi-messenger astronomy community, enabling rapid candidate classification, ranking, and follow-up of newly discovered Multimessenger events.

Simulation-Based Inference Kilonova Modeling Gravitational Waves Binary Neutron Stars Computer Vision LLMs
Explore Research Publications
Phelipe Darc
NS
BH
BH
CBPF · RIO DE JANEIRO
CIERA · NORTHWESTERN 2025–26
NEURIPS 2023 / 2024 / 2025
Scroll
* * *
01

Research

13+
Publications
3
NeurIPS Papers
3
Observatories
GW
Multi-Messenger
Open-Source Tools · MMA Community
KilonovaSCORER — Real-time Kilonova Ranking for GW Follow-up

An open-source framework for scoring optical transient candidates against a kilonova model grid in real time, deployed into TROVE (Tool for Rapid Object Vetting and Examination) — a community coordination platform for multimessenger follow-up. Uses ABC diagnostics and inverse-variance weighting to reject supernova contaminants within 3–4 days of a GW trigger under LSST ToO strategies. Validated on AT 2017gfo and SN 2025ulz.

Inference Tooling · SBI for Time-Domain
Simulation-Based Inference for Kilonova & Transient Parameter Estimation

Building amortized inference pipelines — Amortized Neural Posterior Estimation, Truncated Sequential NPE, Neural Likelihood Estimation — that reduce kilonova parameter inference from hours of MCMC to seconds. Validated on AT 2017gfo and applied to GRB 230307A multi-wavelength data. Symbolic regression methods have extracted analytic scaling laws from BNS merger simulations (NeurIPS 2025).

EM Follow-up · BBH Mergers in AGN Disks
Optical Follow-up of Binary Black Hole Merger Candidates S231206cc & S240413p

Leading multi-epoch optical and spectroscopic campaigns for LIGO/Virgo/KAGRA BBH merger candidates, testing whether remnant interactions with AGN accretion disks produce detectable EM emission. Published in Physical Review D (2025) and arXiv (2026), these are among the most comprehensive optical datasets yet for BBH EM counterpart searches.

LAB-IA · CBPF Collaboration
ADSS: Intelligent Agents & Physical Modeling for the LSST Era

As part of the CBPF LAB-IA group, developing the Astronomical Data Smart System — AI-powered autonomous agents for real-time LSST alert triage (up to 10 million alerts/night), LLM-assisted astrophysical reasoning, and fast SBI physical modeling targeting kilonovae, TDEs, FBOTs, ultra-stripped supernovae, and orphan GRB afterglows.

COIN · Cosmostatistics Initiative — Worldwide Collaboration
Data-Driven Astrophysics with the Cosmostatistics Initiative

Member of COIN — a worldwide interdisciplinary network building data-driven solutions across astrophysics. Current COIN contribution: the SAGUI framework for SED-based segmentation of multi-band galaxy images, applied to JWST/JADES data in GOODS-South (arXiv 2026). COIN projects span statistical methodology, machine learning, and open-source tool development for the global astronomy community.

* * *
02

Publications

Google Scholar ResearchGate ORCID
First Author entries marked in blue  ·  bold = P. Darc
2026
PASP · Submitted First Author CIERA / Northwestern
KilonovaSCORER: Real-time Ranking of Optical Transient Candidates for Gravitational-Wave Follow-up
P. Darc et al.  ·  Developed at CIERA, Northwestern University
Open-source framework for scoring and ranking kilonova candidates from sparse early-time multi-band photometry. Uses per-observation metrics Ptail,KNe and Pnear,KNe aggregated via inverse-variance weighting in logit space, with a sequential ABC diagnostic for temporal consistency. Validated on AT 2017gfo and SN 2025ulz; median supernova contaminant scores fall to zero by 3–4 d post-trigger under LSST ToO simulations while kilonova medians remain ≥ 0.4.
GitHub · phelipedarc/KilonovaSCORER
2026
arXiv Preprint First Author
Searching for Binary Black Hole Merger Emission in AGN Disks: Optical and Spectroscopic Follow-up of S240413p
P. Darc, Bom C.R., Santos A., Panda S., Rodríguez-Ramírez J.C., Kilpatrick C.D., de Oliveira C.M., et al.
arXiv:2603.26972
2026
arXiv Preprint · Instrumentation & Methods
SAGUI: SED-based Segmentation of Multi-band Galaxy Images — Application to JADES in GOODS-South
de Souza R.S., Wille A., Shenoy S., …, P. Darc, Durgesh R.
arXiv:2604.18812
2026
The Astrophysical Journal · Vol 998
SN 2022acko and the Properties of Its Red Supergiant Progenitor: Direct Detection, Light Curves, and Nebular Spectroscopy
Teixeira G., Kilpatrick C.D., Bom C.R., Santos A., P. Darc, et al.
ApJ 998(1):123
2026
arXiv Preprint · High Energy Astrophysics
Search For a Counterpart to the Subsolar Mass Gravitational Wave Candidate S251112cm
Vieira N., Franz N., Subrayan B., Kilpatrick C.D., …, P. Darc, et al.
arXiv:2603.17009
2025
Physical Review D · Vol 112(6) First Author
Long-term Optical Follow-up of S231206cc: Multimodel Constraints on BBH Merger Emission in AGN Disks
P. Darc, Bom C.R., Kilpatrick C.D., Santos A.S., Fraga B., Rodríguez-Ramírez J.C., Coulter D.A., et al.
Phys. Rev. D 112, 6
2025
The Astrophysical Journal Letters · Vol 994(2)
Optimizing Kilonova Searches: A Case Study of the Type IIb SN 2025ulz in the Localization Volume of Gravitational Wave Event S250818k
Franz N., Subrayan B., Kilpatrick C.D., …, Bom C.R., P. Darc, et al.
ApJL 994(2):L45
2025
NeurIPS 2025 · ML and Physical Sciences Workshop First Author
Symbolic Regression Is All You Need: From Simulations to Scaling Laws in Binary Neutron Star Mergers
P. Darc, Bom C.R., Kilpatrick C., Fraga B.M.O., Teixeira G.S.M.
arXiv:2511.08784   Camera-Ready PDF
2024
The Astrophysical Journal · Vol 971(1) First Author
Kilonova Spectral Inverse Modelling with Simulation-Based Inference: An Amortized Neural Posterior Estimation Analysis
P. Darc, Bom C.R., Fraga B., Kilpatrick C.D.
ApJ 971(1):82   arXiv:2311.09471
2024
NeurIPS 2024 · ML and Physical Sciences Workshop First Author
Multi-Wavelength Analysis of Kilonova Associated with GRB 230307A: Accelerated Parameter Estimation and Model Selection Through Likelihood-Free Inference
P. Darc, Bom C.R., Teixeira G.S.M., Kilpatrick C., Sherman N.
Workshop Paper PDF
2024
Monthly Notices of the Royal Astronomical Society · Vol 535(1)
A Dark Standard Siren Measurement of the Hubble Constant Following LIGO/Virgo/KAGRA O4a and Previous Runs
Bom C.R., Alfradique V., Palmese A., Teixeira G., Santana-Silva L., Santos A., P. Darc
MNRAS 535(1):961–975
2024
Monthly Notices of the Royal Astronomical Society · Vol 529(1)
The S-PLUS Transient Extension Program: Imaging Pipeline, Transient Identification, and Survey Optimization for Multi-Messenger Astronomy
Santos A., Kilpatrick C.D., Bom C.R., P. Darc, Herpich F.R., et al.
MNRAS 529(1):59–73
2024
Astronomy and Computing
Photometric Redshifts Probability Density Estimation from Recurrent Neural Networks in the DECam Local Volume Exploration Survey Data Release 2
Teixeira G., Bom C.R., Santana-Silva L., Fraga B.M.O., P. Darc, et al.
A&C 100886
* * *
03

Curriculum Vitae

Education
2023 — PRESENT
Ph.D. in Physics
Centro Brasileiro De Pesquisas Físicas (CBPF) · Rio de Janeiro
Supervisor: Clecio R. De Bom
2021 — 2023
M.Sc. in Physics
Centro Brasileiro De Pesquisas Físicas (CBPF) · Rio de Janeiro
2016 — 2021
B.Sc. in Physics
Universidade Federal Fluminense (UFF)
Selected Talks & Posters
NEURIPS 2025
Symbolic Regression Is All You Need: From Simulations to Scaling Laws in Binary Neutron Star Mergers
ML and the Physical Sciences Workshop · San Diego, USA
NEURIPS 2024
Multi-Wavelength Analysis of Kilonova Associated with GRB 230307A
ML and the Physical Sciences Workshop · Vancouver, Canada
NEURIPS 2023
Kilonova Spectral Inverse Modelling with Simulation-Based Inference
ML and the Physical Sciences Workshop · New Orleans, USA
Experience
DEC 2025 — MAY 2026
Visiting Researcher
CIERA · Northwestern University · Evanston, IL, USA
Developed and submitted KilonovaSCORER to PASP — real-time kilonova ranking framework for GW follow-up using ABC diagnostics and LSST ToO simulations
2022
Research Internship
Laboratório Nacional de Computação Científica (LNCC)
NLP / BERTimbau / Medical Portuguese
2020 — 2021
Research Assistant
DESGW Collaboration — Electromagnetic follow-up of GW events
Technical Skills
Machine Learning
Simulation-Based Inference
CNNs / RNNs / Autoencoders
Bayesian Inference / MCMC
Transformers / BERT
Tools & Languages
Python / PyTorch / JAX
swyft / lampe / numpyro
Astropy / KilonovaNet
Full Astro CV AI / ML CV Master's Thesis
04 / Contact
Open to postdoc and research collaborations

I am currently a Ph.D. candidate at CBPF, actively seeking postdoctoral opportunities in astrophysics, multi-messenger astronomy, or machine learning applied to physical sciences. I welcome opportunities to collaborate on challenging open problems at the intersection of AI and astrophysics.

Email phelipedarc@gmail.com GitHub github.com/phelipedarc LinkedIn linkedin.com/in/phelipedarc Scholar Google Scholar Profile RGate ResearchGate Profile ORCID 0000-0001-8833-474X Kaggle phelipeantoniedarc
Affiliation
Centro Brasileiro De Pesquisas Físicas (CBPF)
Rio de Janeiro, Brazil