CV
Isabel Agostino
Ph.D. Candidate in Industrial Engineering & Operations Research
Berkeley, CA
📧 iagostino@berkeley.edu
🌐 https://iagostino.github.io
LinkedIn: isabelagostino | GitHub: iagostino
Experience
NASA Ames Research Center, Intelligent Systems Division
OSTEM Intern
2026 (Summer) — Mountain View, CA
- Developed a path-planning module for the fmdtools Python library, under review for inclusion in the next release.
- Integrated route-generation methods into existing workflows to enhance autonomous-system simulation under degraded conditions.
- Authored unit and integration tests to ensure numerical robustness and library compatibility, and prepared end‑user examples to support correct and transparent usage.
West Monroe Partners
Consultant & Business Analyst
2022–2023 — Tysons, VA
- Automated financial reporting workflows by developing data processing and validation pipelines, significantly reducing manual error rates and turnaround time.
- Designed structured data querying and validation tools to improve reliability, traceability, and reproducibility of analytical outputs.
- Standardized and documented data workflows to enhance transparency and ensure consistent execution across internal teams and client stakeholders.
- Delivered user-oriented documentation and process guides to support adoption and long-term maintainability of automated tools.
National Institute of Standards and Technology (NIST)
Summer Undergraduate Research Fellow
Summer 2021 — Gaithersburg, MD
- Designed and implemented a scalable genetic algorithm to approximate star discrepancy in high-dimensional quasi-random sequences (an NP-hard optimization problem).
- Analyzed convergence behavior and algorithm efficiency compared to theoretical and exisiting methods..
Education
University of California, Berkeley
Ph.D., Industrial Engineering & Operations Research
2024–2028 (expected) — Berkeley, CA
- Advisor: Dr. Thibaut Mastrolia
- Formulated and solved a singular-stopping stochastic control problem with self-exciting (Hawkes) jump dynamics via a variational HJB equation with gradient constraints.
- Built a discrete-time Markov chain approximation (with a novel Hawkes-process discretization) and proved its value function converges to the continuous-time control problem, then validated the theory with Monte Carlo simulations in Python.
- Developing BSDE/2BSDE techniques for mean-field contract theory under common noise and jump risk, including new Itô calculus for two coupled flows of conditional measures.
University of California, Berkeley
M.S., Industrial Engineering & Operations Research
2023–2024 — Berkeley, CA
- Worked with Dr. Alper Atamtürk on Resilient Logistics (U.S. Department of Defense, Office of Naval Research).
- Designed and implemented large-scale, constraint-intensive optimization models for multi-stage supply chain routing and scheduling; incorporated operational, capacity, timing, and resource constraints across distributed logistics networks.
- Developed stochastic and scenario-based modeling components to account for uncertainty in demand, transit times, and system disruptions.
- Contributed to the development of a deployable decision-support tool, including visualization and reporting components, to support data-driven logistics planning for end users.
William & Mary
B.S., Mathematics (Minor: Economics)
2018–2022 — Williamsburg, VA
- Thesis: Approximating Star-Discrepancy with a Genetic Algorithm
- Graduated summa cum laude and Phi Beta Kappa with departmental honors.
Selected Honors
- Outstanding Graduate Student Instructor Award (2026)
- UC Berkeley Chancellor’s Fellowship (2024)
- Phi Beta Kappa, Alpha Chapter of Virginia (2022)
- Departmental Honors, William & Mary Department of Mathematics (2022)
- James Monroe Scholars Program (2018)
Skills
Languages: Python, Julia, R, SQL, MATLAB
ML Frameworks: PyTorch, scikit-learn, TensorFlow
Publications
Talks
Approximation of Singular-Stopping Control Driven by Hawkes Processes via Rescaled MDPs
Conference Minisyposium at XIII Bachelier World Congress, University of Bologna, Bologna, Emilia-Romagna, Italy
Unifying discrete and continuous time mixed optimal stopping-singular stochastic optimization for power plant investment
Conference Minisyposium at SIAM Conference on Financial Mathematics and Engineering (FM25), Miami, FL, USA