Skip to content
Barry Zhang, home
Open to new rolesIrvine, CaliforniaPacific Time

Machine Learning Engineer · Statistics and research

I take models to production and own what happens next.

I’m Barry (Zehui) Zhang, a Machine Learning Engineer with a statistician’s training. I run sole-author research on heavy-tailed healthcare costs, say plainly what does not reach significance, and took a Bitcoin model through a locked test to an honest negative result.

Tools: PyTorch · scikit-learn · LightGBM · pandas · NumPy · SQL · pytest

See the work ResumeEmail LinkedIn
Headline result · sole-author research
2.0×

the high-cost spending captured at top-0.5% capacity, against a calibrated logistic baseline (p = 0.018)

0.5%1%2%5%10%SPENDING CAPTURED, BY CAPACITY
0
quantile crossings in 745,560 pairs, against 1.54% for baselines
0.006
expected calibration error
98 / 98
manuscript numbers machine-verified against run outputs
Read about the research
Most recently
CVS Health / Aetna
Machine Learning Engineer, Feb 2025 – Nov 2026
Education
M.S. Biostatistics and Data Science
Weill Cornell Medicine
B.S. Joint Mathematics-Economics
UC San Diego
Research
Sole author
5 manuscripts in preparation, 3 patent applications pending
Available
Immediately
Remote, or relocating anywhere in the US
01
Machine learning engineering, end to end

The whole loop, not one part of it.

  1. 01

    Data

    In production, at CVS Health/Aetna

    Nightly pipelines that rebuild the tables the decisions are served from, a reference data product refreshed monthly, and the staging SQL that prepares each day's flagged cases.

    Read the case
    In my own research

    72,818 person-years of national survey data, with rebuilt inputs byte-identical to the originals.

    Read the research
  2. 02

    Model

    In production, at CVS Health/Aetna

    The decision logic of the review pipeline is a text-matching engine I implemented: pattern extraction over free-text notes, fuzzy entity matching with edit-distance limits, and a three-way answer instead of a keyword hit.

    Read the case
    In my own research

    A multi-task FT-Transformer in PyTorch that predicts expected cost, 7 tail quantiles and nested high-cost risk in one model.

    Read the research
  3. 03

    Evaluate

    In production, at CVS Health/Aetna

    Each new rule became a tested release, and production output was reconciled against reference runs. The monitoring dashboard was recomputed from source data before anyone relied on it.

    Read the case
    In my own research

    Paired bootstrap inference, calibration and a 35-run ablation. What did not reach significance is reported as such.

    Read the research
  4. 04

    Ship

    In production, at CVS Health/Aetna

    A real-time decision service on Kubernetes, released behind automated tests and a load test. A breaking contract change staged until every caller had moved. A guarded promotion path from a rule change to production that became the team's reference pattern.

    Read the case
  5. 05

    Monitor

    In production, at CVS Health/Aetna

    Every request recorded as a structured event. Adoption, decision rate and latency by code path, published every day and reconciled to source data.

    Read the case
  6. 06

    Feed back

    In production, at CVS Health/Aetna

    Nearly all missed decisions traced to specific gaps in coverage, and those went back to be closed. A healthy overall latency was hiding a slow path, so latency was split by code path.

    Read the case

At CVS Health/Aetna I was the end-to-end owner of both systems and wrote most of the code in their repositories. I also owned the team's production services and pipelines across dev, test and prod.

The same work, read as a deployment: from a business problem to a system people use
02
Selected work

Production systems, full products and research.

Careful questions, reproducible answers.

  1. 0.5%1%2%5%10%SPENDING CAPTURED, BY CAPACITY
    Independent research · 2025 – present

    Risk-aware healthcare cost prediction

    Sole-author research on who to reach when capacity is fixed: a multi-task deep-learning model of heavy-tailed healthcare spending.

    2.0×high-cost spending captured at top-0.5% capacity
  2. LOCKED TEST, BALANCED ACCURACY0.50 chance
    Personal project · 2026

    Predicting six-hour Bitcoin moves: a negative result

    A blending ensemble for six-hour BTCUSDT moves, taken through a locked test and a costed backtest. The answer was no.

    0.514balanced accuracy on the locked test, interval includes 0.5
  3. Decision APIreal-timeServing tablesAutomation botsAI assistant
    CVS Health / Aetna · 2025 – 2026

    Real-time decision system

    A decision model, put into production as a real-time service that automation bots and an AI assistant call.

    Real-timedecisions for automation bots and an AI assistant
  4. Staging SQLWorkflowAirflowArchivalexactly-onceMatching engineasync serviceDaily delivery
    CVS Health / Aetna · 2025 – 2026

    Manual-review reduction pipeline

    A text-matching engine and the pipeline around it, which find false-positive review flags in free-text notes and clear them every day.

    Dailydelivery that keeps wrongly flagged cases out of manual review
    • Orchestration
      Airflow 2toAirflow 3
    • CI/CD
      JenkinstoGitHub Actions
    • Cloud
      Old environmenttoNew environment
    CVS Health / Aetna · 2025 – 2026

    Production platform and migrations

    Production services and pipelines across dev, test and prod, kept running while the platform underneath them was replaced.

    Dev to prodownership of production services and pipelines, through platform migrations
  6. Pawline storefront
    Independent project · Nov 2026

    Pawline: a custom pet-portrait gift store

    A complete store where a customer uploads a pet photo, places it on a product and approves a designer's proof.

    117automated tests across Chromium, Firefox and WebKit
  7. Bottle page of the wine platform prototype
    Independent project · Nov 2026

    Wine retail commerce and operations platform

    A bilingual storefront, wholesale ordering and four staff workspaces on one schema, correct under concurrency.

    41tables, with role-based access for four staff roles
  8. Resume

    The whole record, on two pages

    Experience, research, projects, skills and education. Read it here or take the PDF.

    Machine Learning Engineer · web and PDF
  9. This site

    How this site is built and checked

    The design system, the drawings, and the checks every page has to pass before it ships.

    Designed, built and run by me

Figures from my time at CVS Health/Aetna are left out of this site. I am glad to walk through them in conversation.

03
Independent research

Who should a care team reach when it can only reach a few?

Healthcare spending is heavy-tailed: a small number of people account for a large part of the total. Most cost models are tuned for the average. This research asks what a model should predict when capacity is fixed and only the top of the distribution can be acted on.

Read about the research
Headline result
Share of total spending captured by the people flagged
0%10%20%30%40%6.2top 0.5%p = 0.0188.8top 1%p = 0.00914.3top 2%p = 0.02826.6top 5%p = 0.07238.9top 10%p = 0.25
model calibrated logistic baseline not significant

Flagging the top 0.5% of people, the model captures 6.24% of total spending against 3.11% for the baseline (2.01×, p = 0.018).

Count-based top-k% selection, model against a calibrated logistic baseline. Paired cluster bootstrap, B = 2,000.

04
About

An engineer with a statistician’s training.

I am a Machine Learning Engineer in Irvine, California. I interned at CVS Health/Aetna in 2024 and returned full time from 2025 to 2026, building the systems that automate decisions that used to need a human.

I studied Mathematics-Economics at UC San Diego and Biostatistics and Data Science at Weill Cornell Medicine. That mix shows up in the work: I like systems that run in production, and I like being able to say how well they run, with an interval around the answer.

Weill Cornell Medicine, Cornell University
M.S., Biostatistics and Data Science
Dec 2024
University of California, San Diego
B.S., Joint Mathematics-Economics; Minor in Business; Provost Honors (4 quarters)
Jun 2023
Full resume PDF
How I work
  1. 01

    Own the whole path

    Design, pipelines, service, rollout, monitoring and support. The hard problems sit at the handoffs between them, so I would rather hold all of them.

  2. 02

    Measure what changed

    A system is finished when there is a number that shows it works, and that number has been checked against the source data.

  3. 03

    Bring the callers along

    A breaking change is a people problem first. I agree on the contract, stage the change and watch adoption until everyone has moved.

Toolbox
ML & Modeling
PyTorch, TensorFlow/Keras, scikit-learn, XGBoost/LightGBM, Optuna
Serving & MLOps
FastAPI, Pydantic, Docker, Kubernetes (GKE), REST API design, streaming, rate limiting, GitHub Actions, Jenkins, Snyk, SonarQube, pytest, HashiCorp Vault
Data & Cloud
GCP (BigQuery, Airflow on Cloud Composer 2 and 3, GKE, GCS, Cloud Monitoring/Logging, Secret Manager, IAM), AWS
Observability
Grafana, Prometheus/PromQL, custom metrics, structured logging, OpenTelemetry, load testing
Full Stack
TypeScript, React, Next.js, Node.js, PostgreSQL, Prisma, Playwright
Languages
Python, TypeScript, SQL, R, Java, C++, Bash
Tools
Codex, Claude Code, Cursor, Git
Domain
Healthcare data, RPA integration, HIPAA/PHI-safe design