# Cameron Beebe, Ph.D.

Agent-friendly: [cameronbeebe.github.io/cv.md](https://cameronbeebe.github.io/cv.md)

Agent-readable version of the CV PDF: [https://cameronbeebe.github.io/Beebe_Resume_SciPhi.pdf](https://cameronbeebe.github.io/Beebe_Resume_SciPhi.pdf)

Last updated: 2026-10-17

[GitHub](https://github.com/CameronBeebe) · [LinkedIn](https://www.linkedin.com/in/cameronbeebe/) · [ResearchGate](https://www.researchgate.net/profile/Cameron_Beebe) · [Email](mailto:cameron@sciphi.info) · [X](https://x.com/BeebsMemes)

- **AI:** Old world experience with numpy, jupyter, conda, uv, keras. New world LLM experience with Devin, Claude Code, Cursor, Grok Bot, Codex, Jev, DSPy, RLMs.
- **Languages:** English (native) and German.
- First ‘classical’ agent-based systems experience in 2014 (Netlogo), and most recently worked on a multi-agent LLM harness.

## Experience

### Dec. 2025–Oct. 2026 Appy.AI.

AI Product Engineer on a multi-agent harness for Violet: a Chief of Staff that routes to specialist agents with scoped context for real workflows.

Shipped a harness eval dashboard comparing batches of executions across configurations. Decided against an external eval framework (e.g. Braintrust) in favor of in-house solutions. Drove qualitative theory-led improvements at low volume. Matched context to harness mechanics (affordance and ergonomics). Orchestration, steering, context management, skills.

### 2021– The SciPhi Initiative.

My LLC for AI research, consulting, and community outreach.

Pair coding and API integration with LLMs in projects like [`scipfs`](https://github.com/CameronBeebe/scipfs) and [`fungi_fortress`](https://github.com/CameronBeebe/fungi_fortress). Researching capabilities, responsible utilization, and limitations of LLMs.

Leading discussion sessions on AI and society at Nederland Philosophy Club (co-organizer).

Interpreting wave mechanics and quantum probabilities as ideal structural objects: [⟨sci|phi⟩i](https://github.com/CameronBeebe/Structural_Idealism)

### 2023–2025 Colorado Outdoors.

Snowboard instructor at Eldora Mountain Resort. Heavy equipment operator at High Performance Earthworks.

### 2015–2021 PhD at GSN (LMU Munich).

AI, cybernetics, and philosophy of science in a structured Ph.D. program at the Graduate School of Systemic Neurosciences.

Dissertation on [Knowledge Transfer in Cognitive Systems Theory](https://edoc.ub.uni-muenchen.de/28655/). Transfer learning in artificial neural networks as effective transfer of control, related to model-based knowledge transfer in cognition. Supported Ashby’s structural view of systems (cybernetics), and a model-based view of computation for diverse computational devices. Took a quantum information course from the physics department, and critiqued quantum-like models of cognition.

Affiliated with the Munich Center for Mathematical Philosophy and the Research Center for Neurophilosophy and Ethics of Neurosciences during this time. Presented work internationally.

### 2019 Deep Learning Intern at Cray, Inc.

Worked for Cray’s Distributed Deep Learning Plugin team. Applied, tested, and experimented with the Plugin across multiple GPUs (gradient averaging using MPI), on a variety of architectures including: LSTM, CNN, RL/DQN, CapsNet. Worked mainly with Keras, scheduled jobs via Slurm, managed ML/DL packages and versions with Anaconda and pip, and committed code to internal git repos. (**Python, Jupyter, Slurm, Linux, Keras, Anaconda, Bash, Git**).

## Education

### 2021 Ph.D. from [Graduate School of Systemic Neurosciences](http://www.gsn.uni-muenchen.de/studies/phd_program/index.html), LMU Munich.

- Dissertation: [*Knowledge Transfer in Cognitive Systems Theory*](https://edoc.ub.uni-muenchen.de/28655/)
- Committee: [Stephan Hartmann](http://www.stephanhartmann.org), [Christian Leibold](https://www.neurotheory.uni-freiburg.de), [Ulrike Hahn](http://www.mcmp.philosophie.uni-muenchen.de/people/faculty/hahn_ulrike/index.html), [Gregory Wheeler](http://gregorywheeler.org)
- Reviewer: [Peter Asaro](https://peterasaro.org/)

### 2015 Masters in [Logic and Philosophy of Science](http://www.mcmp.philosophie.uni-muenchen.de/students/ma/index.html) from [MCMP](http://www.mcmp.philosophie.uni-muenchen.de/index.html), LMU Munich.

- Supervisor: [Karim Thébault](http://www.karimthebault.org), Reviewer: [Michael Cuffaro](http://michaelcuffaro.com/)
- Thesis: *Fluid Mechanical Models in Physical and Computational Contexts*
- Program in mathematical philosophy and philosophy of science including courses on: confirmation theory, formal epistemology, Bayesian networks, probability theory, logic, information theory, and agent based modeling (**Netlogo**).
- Specialized in philosophy of physics.

### 2012 B.A. in Philosophy from Montana State University.

- Thesis: *Uncertainty in Quantum Theory*
- Advisor: [Prasanta Bandyopadhyay](https://www.montana.edu/history/directory/1523887/prasanta-bandyopadhyay)

## Projects (selected)

- **2026 Latest** See latest updates and current projects on [GitHub](https://github.com/CameronBeebe) & [ResearchGate](https://www.researchgate.net/profile/Cameron_Beebe)
- **2018 Project/Talk** *Black Boxes or Rube Goldberg Machines? ANNs as Ashby Regulators.* [Epistemic Opacity in Computer Simulation and Machine Learning](https://www.hlrs.de/events/archive/detail-view/2018-11-28-sas/), HLRS. [(Related Repo)](https://github.com/CameronBeebe/Cybernetics)
- **2018 Project/Talk** Transfer Learning in Artificial Neural Networks: [Introductory Tutorial to Transfer Learning with **Keras**](https://github.com/CameronBeebe/transfer_keras) (**Jupyter** Notebook for PyData Meet-up)
- **2014 Project** [Modeling Memory in Signaling Games](https://www.researchgate.net/publication/321228909_Modeling_Memory_in_Signaling_Games_Deep_and_Surface_Forgetting) (Reinforcement learning in **Netlogo** w/ Dropout)

## Publications

- [Model-Based Computation](https://link.springer.com/article/10.1007/s11047-017-9643-0). Natural Computing, Volume 17, Issue 2, p. 271-281. 2018.
- [Sequent Calculus Representations for Quantum Circuits](https://arxiv.org/abs/1606.06799). Electronic Proceedings in Theoretical Computer Science, volume 214, p. 3-15. 2016.

### WIP Manuscripts (More on my [ResearchGate Profile.](https://www.researchgate.net/profile/Cameron_Beebe))

- [Artificial Neural Networks as Cybernetic Regulators](https://www.researchgate.net/publication/337915649_Artificial_Neural_Networks_as_Cybernetic_Regulators). (Forthcoming)
- [Transfer Learning: The Logic of Discovery for Artificial General Intelligence](https://www.researchgate.net/publication/324330770_Transfer_Learning_The_Logic_of_Discovery_for_General_Artificial_Intelligence)
- [Bayesian Confirmation from Analog Models](https://www.researchgate.net/publication/320259541_Bayesian_Confirmation_from_Analog_Models_Submitted). With [Roland Poellinger](https://www.mcmp.philosophie.uni-muenchen.de/people/external_members/roland_poellinger/index.html).
- [Decoherence and Survival: Why Incompatible Observables Should be Compatible.](https://www.researchgate.net/publication/331983846_Decoherence_and_Survival_Why_Incompatible_Observables_Should_be_Compatible)