MACHINE LEARNING SYSTEMS·BUILD 2026.05·SHIPPING

PHILIP FELIZARTA^PF-01

LEAD MACHINE LEARNING ENGINEER/B2U STORAGE SOLUTIONS

MARK / HISTOGRAM
PMF · BID DENSITY

Philip is the developer of B2U's Bid Engine, the AI-driven trading system that runs 50 MWh of battery storage in CAISO and is scaling to 100 MWh in ERCOT. He builds the ML pipelines around it: price forecasting, asset degradation, incident detection, and financial risk.

Background: applied math and CS at UC Merced, MS in Applied AI at USD. Before B2U, he worked on deep learning for audio at Meets The Eye Studios.

PF/01

PROFILE / 2026 EDITION / PERSONAL DOSSIER

SECTION 01
FIELD RECORD

Experience.

EX-LOG
2022 - PRESENT

ACTIVE
AUG 2023 - PRESENT
LANCASTER, CA

Lead Machine Learning Engineer

B2U Storage Solutions

Lead developer of B2U's Bid Engine, an AI-driven trading system managing 50 MWh in CAISO and scaling to 100 MWh in ERCOT. Builds core ML pipelines for asset degradation, incident detection, and financial risk modeling.
Leading agentification of B2U: building the internal AI agent layer to take over the research, monitoring, and reporting work the team currently does by hand.
Own Spark pipeline development in Microsoft Fabric to maintain large-scale battery and market datasets.
Manage frontend tooling for BI and reporting across Power BI, Streamlit, and React; contribute to daily data workflows across market, battery, and site operations.

CAISOERCOTXGBOOSTTENSORFLOWSPARKFABRICLLM AGENTSREACT

DEC 2022 - AUG 2023
SAN CARLOS, CA

Deep Learning & Software Engineer

Meets The Eye Studios

Designed and implemented deep-learning models (Transformers, ResNets, LSTMs) for speech and audio enhancement using Python and TensorFlow.
Focused on end-to-end prototyping, testing, and deployment of audio enhancement pipelines.

TRANSFORMERSRESNETSLSTMSTENSORFLOW

SECTION 02
SELECTED BUILDS

Projects.

PR-LOG
CAPSTONE / OPTIONS

PR-001 · CAPSTONE

Deep Learning Chess Engine & MaxEnt Nash Optimizer

Built a convolutional + transformer-based chess engine trained with custom self-play data generation. Used Bayesian active learning for hyperparameter tuning across multiple MCTS configurations.

Implemented a max-entropy Nash equilibrium optimizer to construct meta-policies over agents during training.

TRANSFORMERSMCTSSELF-PLAYBAYESIAN OPT.NASH EQ.

PR-002 · QUANT

Implied Stock PMFs & Discrete Portfolio Optimization

React + TypeScript app for options analytics: macro “Due Diligence” dashboard with historical distributions, interactive histograms, and options chains. Custom React visualizations for multimodal price distributions.

Inverted European call/put prices into implied discrete probability distributions via a discretize-then-optimize framework. LLM integration converts written market views into quantitative shifts of the implied distribution.

Applied L-BFGS on model-independent pricing equations; tested on VIX/SPX to highlight crashophobic sentiment. Built omega-ratio constrained optimizer modeling asset dependencies with empirical multivariate copulas.

REACTTYPESCRIPTL-BFGSCOPULASVAR / ESLLM

SECTION 03
CAPABILITIES

Stack.

SK-LOG
ACTIVE STACK

LANGUAGES

Python · JavaScript · TypeScript · C / C++ · Java · R · SQL

FRAMEWORKS

React · TensorFlow · Keras · PySpark

ML / DATA SCIENCE

Machine Learning · Deep Learning · Reinforcement Learning

MATH & ANALYTICS

Mathematical Modeling · Statistics · Optimization · Numerical Methods

TOOLS & PLATFORMS

Power BI · MATLAB · Git · Linux · L^AT_EX

FINANCE

Risk Management · Financial Derivatives · Option Pricing · Bond Pricing

SECTION 04
TRAINING

Education.

ED-LOG
USD / UCM

ED-001

M.S. Applied Artificial Intelligence

University of San Diego

AUG 2023 - MAY 2025

ED-002

B.S. Applied Mathematics & B.S. Computer Science

University of California, Merced

CONFERRED AUG 2022 · DATA SCIENCE CONCENTRATION