Levi Neuwirth — Vita
The formal record. For a less formal, more detailed introduction to who I am, see Me; for what I am actively working on this month, see Current.
Documents
- Curriculum Vitae (PDF) — the complete record, including grants, affiliations, languages, and technical skills.
- Resume (PDF) — one page, engineering-facing.
- Project index — engineering artifacts in depth, with links to writeups and source.
The sections below are generated from the same data as the two PDFs, so they cannot fall out of step with them.
Research Interests
My work clusters into four threads:
- AI safety and applied AI — zero-knowledge proofs for cryptographic verification of large language models, as a MARS V fellow with the Cambridge AI Safety Hub (mentored by James Petrie, Future of Life Institute); a Magic: The Gathering reinforcement-learning project; and reasoning, evaluation, and red-teaming research contracts.
- Mathematics — graph theory, number theory, and theoretical computer science. Public results so far are graph-theory-centered: static coverage and persistence in tree-ball geometry (preprint), and the annealed critical window for growing-radius domination in random regular graphs (preprint).
- Computer systems and high-performance computing — the Weenix kernel, a TCP/IP networking stack from scratch in Go, and micro-architectural performance work (SIMD, hardware counters via PAPI, RAPL energy, cross-ISA ports across AVX2 / ARM NEON-SVE / RISC-V V) on Brown’s OSCAR HPC cluster.
- Machine learning — order-invariant ICD-10-CM embeddings (under review at JAMIA, deployed calculator), the NeuroPose 3D-kinematics system in Liqi Shu’s lab at Brown Neurology, and ongoing research engineering at NeuroAI. Undergraduate work has been clinically focused; graduate study broadens the scope.
Computer vision and security thread through all four but do not stand on their own.
Education
Technical University of Denmark (DTU)
September 2026 – expected 2028MSc in Computer Science and Engineering · Kongens Lyngby, Denmark
Intended path toward PhD upon completion.
Brown University
August 2022 – May 2026Sc.B., Mathematics and Computer Science · GPA 3.8/4.0 · Providence, RI
Primary undergraduate advisor: D. Ellis Hershkowitz (Computer Science). Primary research supervisor: Liqi Shu, MD (Neurology, Warren Alpert Medical School).
DIS Copenhagen / Københavns Universitet
Fall 2024Semester abroad · Copenhagen, Denmark
Research Experience
Cambridge AI Safety Hub
July 2026 – October 2026MARS V Fellow · Cambridge, United Kingdom
- Selected for MARS V, a competitive part-time AI-safety research fellowship, working under James Petrie (Future of Life Institute).
- Research on zero-knowledge proofs for large language models, toward cryptographic verification of claims about model training, inference, and deployment.
Shu Laboratory, Brown University Department of Neurology
October 2023 – May 2026Undergraduate Researcher and Technical Lead · Providence, RI
Research with Dr. Liqi Shu spanning clinical machine learning, movement-disorder kinematic analysis, and claims-based outcome prediction. Two distinct research threads:
- Technical lead on NeuroPose, a 3D pose-estimation and kinematic-analysis system for neurological-recovery research. Built end-to-end pipeline: Python/TensorFlow deep-learning inference, MATLAB statistical post-processing, Rust backend with HTML/JS frontends. System supports four externally-funded sub-projects; clinical-implications manuscript in preparation.
- Co-lead developer on an order-invariant ICD-10-CM embedding model for 30-day readmission (AUC 0.75 vs 0.66 CCI baseline) and postdischarge mortality prediction (AUC 0.86 vs 0.78), Deep Sets architecture trained on 113M+ adult hospitalizations from the HCUP Nationwide Readmissions Database. Preprint under review at JAMIA; public calculator deployed at levineuwirth.github.io.
Shu Laboratory, Brown University Department of Neurology
May 2026 – PresentResearch Affiliate · Providence, RI
Brown University Center for Computation and Visualization (CCV)
March 2025 – July 2026Independent Researcher · Providence, RI
- Micro-architectural study of SIMD contributions to lattice-based post-quantum cryptography (ML-KEM / Kyber) on x86-64 AVX2, conducted on Brown's OSCAR HPC cluster with staff support. Phase 1 released as a technical report (see Publications).
Industry Experience
Independent Research Contractor
2025 – PresentFrontier AI labs
- Code and mathematics contributions across agentic workflow design, task and capability evaluation, reasoning assessment, model safety, and red-teaming.
NeuroAI LLC
March 2026 – Present- Early-stage clinical-AI venture. Leading model development and deployment infrastructure.
Publications and Preprints
Beyond Comorbidity Indices: An Order-Invariant ICD-10-CM Embedding for Readmission and Mortality Prediction
2026Under review at Journal of the American Medical Informatics Association (JAMIA)
The Annealed Critical Window for Growing-Radius Domination in Random Regular Graphs
2026, JulyPreprint
NeuroPose clinical-implications manuscript
2026–2027In preparation. Shu laboratory, Brown University Department of Neurology. Target venue TBD
† Equal contribution among undergraduate authors.
Presentations
Order-Invariant ICD-10-CM Embedding for Readmission and Mortality Prediction: Toward Multimodal Generative Patient Models
August 2026IEEE/ACM Conference on Connected Health: Applications, Systems and Engineering Technologies (CHASE)Presented
Early Detection of Neurological Disorders through Video-Captured Kinematic Analysis
August 2025Poster presentation, Summer UTRA Symposium, Brown University / Rhode Island Hospital Neurology Department
Proving What a Datacenter Did: Verified Inference Between Adversaries
July 2026MARS V, Cambridge AI Safety HubPresented