Working systems with measured results: a planning agent for farm management that runs in the browser, a knee-MRI model for clinical imaging taken apart and reproduced, and a research plan for 3-D anatomical scene models.
A MuZero macrocontroller that plans each farm day over ten strategic plans, with a hemisphere crew layer and a Hungarian-assignment microcontroller below it. You can replay its games and re-run its tree search in the browser. T...
RSNA's 2026 knee-MRI challenge scores twelve findings by mean AUC. I took the 0.956/0.941 public tri-model blend apart down to tensor level and reproduced it exactly. I then tested three changes whose weights were fixed before ...
A research plan for a fine-tuned LLM that reads, repairs and writes OpenUSD (.usda) scenes. It is anchored on my Archetypal Knee USD model and scored with a layered metric, usdfid. The page covers the datasources and their lice...
All notebooks and datasets on Kaggle → · Toolkit: cheat sheet, Evidently template, Llama agents →
Texas Tech Seminar Publication and Peer Review Feedback
I give senior scientific professionals critical feedback on their talks: questions, peer review of their claims and statements, and an annotated inline commentary and critical analysis of what they presented.
The talk presents embedding-based routing at three levels: converting dense LLMs into mixture-of-experts models, routing prompts to sub-networks of a diffusion model or to models in an LLM pool, and routing diffusion ...
The talk presents Realm's five components (tasks, events, the machine model, data and the DMA engine) using one tensor linear layer, then shows how cuPyNumeric runs NumPy programs on Realm and scales a microscopy reco...
The talk covered in-home wearable sensing of older adults, self-taught and contrastive multitask learning for activity recognition, an impairment indicator built from model errors, and small vision-language models run...
Three deployed applications, each improving the daily work of a specific profession, with the checks that make the AI trustworthy listed alongside.
Research agents gather Bell County court, county and state information into one place. That includes a court directory with contacts, judicial scorecards, candidate dossiers and a ballot guide, so people stop piecing it together from dozens of government sites.
A self-hosted Evidently AI server logs agent runs as datasets, versioned prompts and OpenTelemetry traces, alongside an experiment ledger for Kaggle challenges. An agent's quality is then measured run over run instead of judged by eye.
Rebuilds Adobe-only USDA forms as LaTeX and portable fillable PDFs, then links every box that asks for the same thing. A first name typed once fills every name box in the packet, and the county is looked up from a plain-language description of the plot.
Eight recent papers from AAAI, ILRT, IML and IEEE on LLM planning, formal verification and agent trace analysis, with notes on how each relates to verifiable-agent research.
A Trace-Language Framework for Agent Verification
Why textbook RSA preserves multiplicative structure and enables existential forgery, and how hash-then-sign and padding schemes like RSA-PSS close the gap.
How Lagrange and Cauchy shape real-world DH and ECC groups, why nearly-prime cofactor curves trade purity for performance, and what Secure Scuttlebutt teaches about validation.
From Weierstrass curves and double-and-add to cofactor attacks and Dual_EC_DRBG—how subgroup geometry and implementation choices determine ECC trust boundaries.
A structured look at HNP-based key recovery, BIGNUM representation leaks, lazy resizing CVEs, and protocol-level subgroup failures in modern signature stacks.
Bridging prime-order proofs with cofactor reality: small-subgroup attacks, BIGNUM leaks, twist security, and a practical mitigation checklist for protocol engineers.
Why finite-field DLP hits a scalability wall, how ECC and Pollard rho change the cost curve, and a worked finite-field point arithmetic example on F23.
A robust mathematical architecture unifying spatial statistics, duration-explicit sequence modeling, and Bayesian non-parametrics for sub...
A comprehensive hardware-software co-design project that simulates vLLM microarchitectural bottlenecks (KV cache, TLB) and applies those ...
A comprehensive reinforcement learning project implementing Q-Learning algorithm to teach an agent to play Pac-Man, demonstrating fundame...
Questions about cryptography, formal verification, or just want to connect? I'd love to hear from you.
A working notebook — half draft, half archive. The site indexes coursework, project write-ups, and longer pieces on cryptography and verification. Source on GitHub.