Kian Maleki

AI Safety Researcher · Data Scientist · Physics PhD

Kian Maleki, PhD

Scientific computing meets trustworthy AI.

Abstract

I do empirical AI-safety research on reasoning models and build machine-learning and document-intelligence systems, with a research scientist's habit of asking: What can fail, how do we know, and how do we measure it? My work connects physics, careful evaluation, and practical AI tools.

Figure 1. A four-layer feed-forward network under repeated forward passes. Node fill is activation; edge opacity is weight magnitude. Weights are random: an illustration, not a model from the research.

Focus

Three areas, one thread: careful modeling.

My background moves from physics and numerical simulation into applied machine learning and, most recently, empirical AI safety. The connecting idea is the same: build models, test their limits, and explain when they should be trusted.

  • AI Safety & Reasoning Models

    • Chain-of-thought faithfulness
    • answer commitment
    • pre-registered causal experiments on reasoning traces
    • prefix interventions
    • resampling studies
    • evidence about what visible reasoning actually reveals
  • Machine Learning & Model Evaluation

    • Classifier ambiguity
    • calibration
    • threshold-dependent behavior
    • finite-horizon learning-curve prediction
    • conservative extrapolation
    • experiment tracking
    • reproducible benchmarks
  • AI Harnesses & Automation

    • Agent harnesses and orchestration
    • tool use with verification loops
    • automating research workflows
    • industry process automation
    • evaluation and monitoring
    • reproducible pipelines

Selected work

Projects with a research spine.

These projects are designed to be more than portfolio demos. They test reliability, expose failure modes, and connect mathematical ideas to usable AI systems.

  • Reasoning Trace Experiments

    AI Safety
    View project ↗

    A study of where a reasoning model's answer takes shape, resampling its chain of thought at block boundaries.

  • Decided Mid-Thought

    AI Safety
    View project ↗

    A pre-registered causal study of chain-of-thought faithfulness in a reasoning model: prefix interventions and 500 resampled continuations test whether visible reasoning reveals when the final answer becomes determined.

  • Ambiguity Range Framework

    Model Evaluation
    View project ↗

    A diagnostic toolkit for evaluating binary classifiers beyond a single threshold, with ambiguity-focused metrics, calibration analysis, confidence intervals, and statistical testing.

  • Sequence Acceleration Benchmark

    Learning Curves
    View project ↗

    A reproducible benchmark for finite-horizon gradient-boosting learning-curve prediction across many convergence regimes, noise levels, and extrapolation methods.

  • PharmaDoc-AI

    Document AI
    View project ↗

    A pharmaceutical document-intelligence and RAG project that combines PDF extraction, structured information retrieval, document question answering, and evaluation workflows.

  • LumberLex

    Data Normalization
    View project ↗

    A lumber product-name normalization engine that maps inconsistent seller strings to canonical structured outputs using fuzzy matching, dimension extraction, treatment detection, confidence scoring, a Streamlit interface, and chatbot-style explanations.

  • Heart Disease Prediction Pipeline

    MLOps
    View project ↗

    An end-to-end MLOps pipeline with DVC, MLflow, pytest, GitHub Actions, model-performance checks, and drift monitoring with Evidently.

Research

Physics roots, machine-learning branches.

My research background spans quantum materials, crystal-field theory, noncollinear magnons, molecular and defect-mediated transport, dark matter phenomenology, modern ML diagnostics, and, most recently, empirical AI safety on reasoning models.

  1. [1]
    Nano Letters, 2025. A General and Modular Approach to Solid-State Integration of Zero-Dimensional Quantum Systems. Nano Lett. 25, 13787–13794. doi:10.1021/acs.nanolett.5c03125
  2. [2]
    Physical Review B, 2025. Crystal fields, exchange and dipolar interactions, and noncollinear magnons of erbium oxide. Phys. Rev. B 112, 064434. doi:10.1103/v6n3-srt5
(1)
The Er₂O₃ Hamiltonian of [2]: dipolar, exchange and Zeeman terms for the erbium moments.
(2)
Equations of motion for the magnon operators in [2]. The negative sign on the right of the second equation is why ordinary diagonalization is not enough.
(3)
The paraunitary condition used in [2] to obtain the 32 magnon branches while preserving the bosonic commutation relations.

Research interests

  • AI safety
  • Reasoning models
  • CoT faithfulness
  • Classifier ambiguity
  • Learning curves
  • Calibration
  • Scientific ML
  • Document intelligence
  • RAG
  • High-dimensional modeling
  • Numerical simulation

Experience

A bridge from research to applied AI.

My path combines academic research, teaching, scientific software, and project-based machine-learning training.

201520162017201820192020202120222023202420252026Independent Researcher · AI automation and LLMsBlueDot Impact · Technical AI Safety ProjectCreighton · Resident Assistant ProfessorTripleTen · AI & ML ProgramPfizer / Extern · Document IntelligenceIowa · Graduate Research AssistantCreighton · Graduate Research Assistant201520172019202120232025Independent Researcher · AI automation and LLMsBlueDot Impact · Technical AI Safety ProjectCreighton · Resident Assistant ProfessorTripleTen · AI & ML ProgramPfizer / Extern · Document IntelligenceIowa · Graduate Research AssistantCreighton · Graduate Research Assistant
Figure 3. Positions, 2015 to present. Solid bars: research and teaching; the arrow marks ongoing work. Outlined bars: training and externship.
  1. 2025 – present

    Independent Researcher

    AI automation and LLMs

    Self-directed, open-source research and development in AI automation and large language models, with public code and write-ups on GitHub.

  2. 2026

    Technical AI Safety Project

    BlueDot Impact

    Completed the program research project, covering Decided Mid-Thought and Reasoning Trace Experiments.

  3. 2025 – 2026

    Resident Assistant Professor of Physics

    Creighton University

    Taught undergraduate physics, developed optics and electronics lab materials, supervised teaching assistants, and integrated AI tools into scientific workflows.

  4. 2025 – 2026

    AI & Machine Learning Training

    TripleTen

    Completed project-based training across supervised learning, unsupervised learning, deep learning, NLP, computer vision, SQL, Git, and end-to-end ML workflows.

  5. 2026

    AI-Powered Document Intelligence Externship

    Pfizer / Extern

    Built retrieval-augmented and document-intelligence workflows for healthcare documents using native PDF extraction, LLMs, and RAG architectures.

  6. 2018 – 2025

    Graduate Research Assistant

    University of Iowa

    Developed high-dimensional numerical models, optimized scientific simulations, analyzed noisy data, and contributed to peer-reviewed work in condensed matter physics and quantum materials.

  7. 2015 – 2018

    Graduate Research Assistant

    Creighton University

    Built simulation-based models using Python and Fortran, including N-body simulation, molecular dynamics, and parallelized scientific computing workflows.

Tools

Technical stack.

Machine Learning & LLMs
Scikit-Learn, XGBoost, PyTorch, Transformers, LLM APIs, reasoning-trace analysis, calibration, model evaluation, NLP, RAG, MLflow.
Engineering
Python, SQL, FastAPI, Pydantic, Docker, Linux/Bash, Git, GitHub Actions, Jenkins, CI/CD, pytest.
Scientific Computing
NumPy, Pandas, Polars, Numba, Fortran, HPC, MPI, numerical optimization, Monte Carlo simulation.
Document Intelligence
PDF parsing, information extraction, OCR-aware workflows, structured extraction, evaluation of extraction quality.