Darrell S. Best Jr.
Hugging Face

Darrell S. Best Jr.

I build AI agents and the models behind them.

Senior AI Research Engineer II at USC ISI. I build agent systems that handle real multi-step work, and I fine-tune, edit, and quantize the open-weight models they run on, published on Hugging Face with the eval numbers attached.

View my models See the agent work

42 open models on Hugging Face

Portrait of Darrell S. Best Jr.
USC ISI, Washington, D.C.
Models on Hugging Face
42
Downloads, last 30 days
49K
Years shipping applied AI
7+
Years building agents
3+

01

Agents that do the actual work

I've been building agents for 3+ years: systems that run long tool loops against messy real environments, remember what went wrong, and back off safely when something breaks.

Software debloating

Agents that build, then shrink, unfamiliar codebases

Debloating only works on code that builds, and every cut risks breaking it. These agents pull source from git, work out the build requirements, containerize the project in Docker, and drive the build loop themselves. Once it builds, they call debloating tools with increasing aggressiveness, roll back to the last working configuration the moment something breaks, and write up what was removed and why.

Related: An Execution Soundness and Security Benchmark for Java Debloating Tools, accepted poster at ACSAC 2026.

  1. Clone the repo
  2. Infer the build
  3. Containerize
  4. Build, fix, repeat
  5. Debloat harder
  6. Roll back on break
  7. Report

Agent memory

ECHO

A memory tool I built so agents keep a record of failed builds and fixes. The next run starts from what already went wrong instead of repeating it.

Healthcare

Clinical coding agents

Agents that read unstructured clinical notes, reason over candidate ICD codes, and produce the mapping, cutting down the manual review that dominates medical coding.

Financial crime

Anti-money-laundering agents

Agentic approaches to spotting laundering patterns across transaction graphs, with output that analysts can read and act on.

02

Models on Hugging Face

Every release ships with its numbers: refusal rate, KL divergence against the base model, reasoning and tool-call checks, plus ready-to-run GGUF, FP8, and NVFP4 builds.

Most downloaded

Qwen-Image-2.1 Prompt Enhancers

Refusal-ablated versions of the text-to-image and image-edit prompt rewriters for Qwen-Image-2.1, a fine-tuned Qwen3.5-VL 9B. Each one was picked from a 1,000-trial Pareto search, checked through a real editing app's prompt-enhancer path, and repacked as single-file ComfyUI builds.

9 repos17K44

T2II2IComfyUIGGUFFP8NVFP4Text encoder
9/100edit refusals (was 100)
0.0368KL, image edit
0.0130KL, text-to-image

Small and fast

Qwen3.5 Heretic series

Edge-friendly 0.6B to 9B models, each evaluated after export and published with quantized builds for llama.cpp and vLLM. The 9B lands at 5/100 refusals.

0.6B0.8B2B4B9B

20 repos22K13

Vision-language

Qwen3.8-27B Heretic

A 31-trial parameter search landed at 0/100 refusals with less drift than the prior public release. Vision, thinking control, and MTP weights intact.

0/100refusals
0.0465KL divergence
bf16GGUFFP8NVFP4

4 repos7.4K

27B writing model

Hemmingway-1 Heretic

A Qwen3.8 27B writing fine-tune with refusals removed by full-weight ARA. Reasoning held at 40/40 on GSM8K, the vision tower is restored, and LoRA tuning still works.

0/100refusals (was 98)
0.083KL divergence
bf16Text-onlyGGUFFP8NVFP4

9 repos2.3K5

Browse all 42 models

Live stats from the Hugging Face Hub.

03

Custom models, fine-tuning runs, and the rest of the pipeline

Each line below comes from a released model or shipped system, with the number that backs it up.

  1. Fine-tuning

    LoRA, QLoRA, DPO, full fine-tunes

    As a fine-tunability check, Hemmingway-1 Heretic got a rank-16 LoRA on persona and tool-call chats. The adapter merged back cleanly and the model answered unseen prompts with well-formed tool calls.

    0.96 → 0.004training loss
  2. Model editing

    Arbitrary-Rank Ablation

    ARA rewrites attention and MLP output projections directly, no retraining. On Qwen3.8-27B, a 31-trial search reached 0/100 refusals with less drift than the previous public release.

    0.0465KL vs. base model
  3. Custom models

    Weight grafts and architecture surgery

    Hemmingway-1 shipped as text-only weights. I grafted it back into the stock Qwen3.8 vision-language layout, restored the multi-token-prediction head, and checked the result against the source.

    1:1logits vs. source
  4. Quantization

    GGUF, FP8, NVFP4

    Releases get llama.cpp builds from Q4_K_M to BF16, plus FP8 and NVFP4 for vLLM on Blackwell. Each build is re-tested for drift, refusals, and tool calls after conversion.

    27 GB27B model in NVFP4
  5. Training at scale

    DeepSpeed and federated learning

    Multi-GPU fine-tunes of multilingual transformers with DeepSpeed, and federated training with Flower for data that can't leave its owner, using aggregation that holds up when one participant misbehaves.

    7+ yrsshipping applied ML

04

Background

I'm a Senior AI Research Engineer II at USC's Information Sciences Institute, based in Washington, D.C. Over seven years I've worked on computer vision for defense sensors, multilingual LLMs, federated learning, clinical NLP, and, for the last three years, agents.

I also release open-weight models on Hugging Face, write on the blog, and I'm finishing an MS in Computer Science at USC.

Where I work

05

Degrees

MS in Computer Science

2024 to present

University of Southern California, Viterbi School of Engineering. Focus: Data Science.

Applied NLP, machine learning, data mining, information retrieval, algorithms, and database systems.

Graduate coursework
  • CSCI 544, Applied NLP: transformer language models, fine-tuning, and LLM pipelines.
  • DSCI 552, ML for Data Science: supervised and unsupervised ML, recommenders.
  • DSCI 553, Data Mining: frequent patterns, LSH, clustering, link analysis, streaming.
  • CSCI 561, Foundations of AI: search, CSPs, probabilistic reasoning, planning, game-playing agents.
  • CSCI 567, Machine Learning: kernel methods, ensembles, deep learning.
  • CSCI 570, Analysis of Algorithms: dynamic programming, graph algorithms, NP-completeness.
  • CSCI 572, Information Retrieval: crawling, indexing, ranking, neural retrieval.
  • CSCI 585, Database Systems: relational and NoSQL, query optimization, distributed data.

BS in Computer Science

2012 to 2017

Clemson University, School of Computing. Minor: Philosophy.

Systems programming, algorithms, and software engineering, plus graduate-level electives in HCI and eye tracking.

Undergraduate coursework
  • CPSC 1010 / 1020: programming in C and C++.
  • CPSC 2120: algorithms and data structures.
  • CPSC 2150: software development foundations.
  • CPSC 2310: computer organization.
  • CPSC 3220: operating systems.
  • CPSC 3500: foundations of computer science.
  • CPSC 3520: programming systems.
  • CPSC 3600: networks and network programming.
  • CPSC 3720: software engineering.
  • CPSC 4620: computer graphics.
  • CPSC 4140 / 6140: human-computer interaction (graduate level, Prof. Andrew Duchowski).
  • CPSC 4120 / 6120: eye tracking methodology (graduate level, Prof. Andrew Duchowski), which led to my ETRA '16 paper.

06

Papers and posters

  1. 2026

    An Execution Soundness and Security Benchmark for Java Debloating Tools

    Accepted poster, IEEE ACSAC 2026, the Annual Computer Security Applications Conference, December 7 to 11, Los Angeles.

    Accepted
  2. 2016

    A Rotary Dial for Gaze-based PIN Entry

    Best, D. S. and Duchowski, A. T. Proceedings of the Ninth Biennial ACM Symposium on Eye Tracking Research & Applications (ETRA '16), pp. 69 to 76. ACM.

    ETRA '16
Google Scholar profile

07

Stack

Training

PyTorchTransformersPEFT / LoRATRLDeepSpeedLightningJAXFlower

Serving and quantization

vLLMllama.cppGGUFFP8NVFP4OllamaComfyUI

Engineering

PythonC++JavaScriptSQLDockerLinuxCI/CD

09

Games

Coming soon. Nothing here is playable or released yet, and no release date is set.

  1. Aeon Ascendant header art: pixel-art title, a crowd of creatures, and a purple-haired character.

    Aeon Ascendant

    Coming soon

    Wishlist on Steam

  2. Coming soon

    Fickle

    No approved short description yet.

    Wishlist on Steam

  3. Coming soon

    City of No Angels

    Post-apocalyptic survival horror set in a 1:1 Los Angeles. Not on Steam yet.

08

Get in touch

Hiring, collaborating, or curious about one of the models? Email is fastest.