We've spent the last few articles building an arsenal.
GitHub for version control and collaboration. Obsidian for organizing your thinking into a linked knowledge graph. Jupyter Notebooks for testing your ideas computationally. LaTeX for publishing results that look like they belong in a journal. And a 13-step research methodology mapped to the Tree of Life that tells you how to think through hard problems from first principles.
Tools without process are toys. Process without tools is theory. This series puts them together.
What This Series Is
The Researcher's Path is a 13-part practical guide to doing real scientific research using AI as your research partner. Not "ask ChatGPT a question and paste the answer into a Word document" research. The kind where you question the question, challenge the assumptions the field takes for granted, test your hypotheses computationally, and produce results that actually contribute to human knowledge.
Each part builds on the previous one. By the end of the series, you'll have:
- A fully configured research environment (Obsidian vault, Jupyter workspace, GitHub repository)
- A trained AI research partner that challenges you instead of agreeing with you
- A complete research project from initial question through computational testing to publication-ready paper
- The methodology to repeat this process for any problem in any field
New parts release every Saturday.
Why I Built This
When I started my quantum theory and engineering research, I had AI. I had the data. I had the tools. What I didn't have was a process for taking an AI conversation about a physics problem and turning it into a testable, publishable piece of science.
I had to build that process through trial and error. Months of fighting with AI models that wanted to give me textbook answers instead of challenging the textbook. Months of figuring out how to organize hundreds of AI conversations into something coherent. Months of learning how to take a mathematical insight and formalize it into something a physicist would take seriously.
Here's what I eventually proved, as a proof of concept: I demonstrated retrocausality as a unifying framework across quantum mechanics and gravity using four independent public datasets: CMS LHC dimuon events, Planck CMB maps, LIGO gravitational wave catalogs, and IceCube DeepCore neutrino data. Not by running new experiments. By analyzing data that was already published, already available, already sitting there waiting for someone to look at it with the right methodology.
The CMS dimuon analysis alone came back at over 1,000 sigma significance for pre/post reaction asymmetry. One thousand. The Planck CMB results exceeded 6 sigma. LIGO waveform anomalies exceeded 5 sigma. IceCube neutrino oscillation deviations exceeded 7 sigma. Every dataset, independently, pointed at the same conclusion: the retrocausal wavefunction model outperforms standard interpretations.
I did this while deliberately presenting myself as a non-academic researcher from a private corporation, working with an AI research partner (Grok), using publicly available data that any grad student on earth could have downloaded. The paper is in the repo. The methodology is what this series teaches.
Let that settle in for a second. Physics has been in a deadlock for decades. Quantum gravity remains "unsolved." The biggest corporations in the world are doing layoffs while sitting on mountains of unanalyzed data. PhD programs are producing graduates who can cite the literature but can't question whether the literature is complete. And meanwhile, the data to bridge quantum mechanics and general relativity was sitting on public servers, waiting for someone to apply a time-symmetric model that's been theoretically valid since Huw Price published on it in 2012.
The credentials didn't matter. The institutional backing didn't matter. The process mattered. That's what this series teaches.
The Series Map
This series is layered like our Developer's Tree and Greek language series: each part maps to a level of the Tree of Life, building from foundational preparation through active analysis to manifestation.
The Researcher's Path: 9 Parts, Every Saturday
The Three Veils (Setting Up)
Part 1: Before You Research (Apr 12) — Ain Setting up your research environment from scratch. Obsidian vault architecture for research projects. Jupyter workspace with the right libraries. GitHub repository structure. The folder hierarchy that keeps a thousand notes organized. You build the lab before you start the experiments.
Part 2: Conditioning Your AI (Apr 19) — Ain Soph How to create system prompts and project contexts that turn AI from a search engine into a research partner. How to make it disagreeable but collaborative. How to take manual snapshots and preserve insights. How to fight past the training bias toward safe, consensus answers. This is the most important skill in the series.
Part 3: The Survey (Apr 26) — Ain Soph Aur Mapping everything that's been tried. Using AI to survey an entire problem space, not just the top results. Identifying where solutions exist, where gaps are, and where multiple fields point at the same unanswered question. Your retrocausality journey starts here: survey the literature and discover what the data actually says versus what people claim it says.
The Upper Tree (Analysis)
Part 4: The Root Question (May 3) — Kether How to strip a problem down to its first-principles question. The difference between the question everyone asks and the question that actually matters. Why "how do we mitigate X?" and "what causes X?" produce radically different research trajectories. Working examples with real problems.
Part 5: Classification and Structure (May 10) — Chokmah + Binah The flash of insight and the framework that captures it. How to identify the actual components of a problem that the consensus view blurs together. How to build a conceptual model before you build a mathematical one. Turning intuition into architecture.
Part 6: Cross-Domain Expansion (May 17) — Chesed Where the missing answers live. Using AI to pull data and methods from adjacent fields. Why the same coefficient operations appear across physics, engineering, and chemistry. Why fragmented domain knowledge hides solutions in plain sight. The step where your research gets genuinely exciting.
The Lower Tree (Testing and Publication)
Part 7: Root-Level Analysis in Practice (May 24) — Geburah + Tiphareth The cut and the balance. Critical analysis of existing explanations. Identifying fitted versus derived coefficients. Solving for unknowns using the algebra of observables. This is where the retrocausality working example comes together: the actual analysis, the methodology that produced >1,000 sigma on CMS data alone, and why a non-academic researcher with the right process outperforms a field that stopped questioning its own assumptions.
Part 8: Formalization and Testing (May 31) — Netzach + Hod + Yesod From insight to mathematics to computation. Writing equations in LaTeX. Testing them in Jupyter. Recording permutations in Obsidian. The discipline of formalization and the power of computational verification. Your notebook becomes your laboratory.
Part 9: Publication and Iteration (Jun 7) — Malkuth Manifesting your research. Writing the paper. Building the reproducible notebook. Pushing to GitHub. Sharing with the community. And the most important step: identifying what the next question is, because Malkuth is never the end. It's the beginning of the next cycle.
What You'll Need
The good news: everything is free.
- Obsidian — Free. Your research knowledge graph.
- Jupyter / Google Colab — Free. Your computational laboratory.
- GitHub — Free. Your version control and collaboration platform.
- Overleaf — Free tier. Your LaTeX editor for publication.
- An AI model — Claude, ChatGPT, GROK, or Gemini. Free tiers available for all.
- Curiosity — The only requirement that can't be downloaded.
If you've read the tool articles, you already have most of this set up. If you haven't, Part 1 walks you through everything from scratch.
How to Follow Along
Each part includes:
- Conceptual explanation of that stage of the methodology
- Hands-on instructions for setting up or using the relevant tools
- Working examples that build toward the retrocausality analysis
- Exercises you can apply to your own research questions
- Downloadable templates (Obsidian vault structures, Jupyter notebooks, system prompts)
You can follow along with the retrocausality example or bring your own research question. The methodology works for any field, any problem, any level of expertise. That's the point.
The Bigger Picture
This series isn't just about learning to do research. It's about building toward something larger.
The tools and methodology we're covering here are the foundation for a universal research planning platform. A system where researchers across domains can share methodologies, permute data discoveries, and collaborate on problems that no single field can solve alone. Where the gaps between physics and chemistry and biology and engineering stop being barriers and start being opportunities.
We'll need to consider resource allocation, production design, and scalable infrastructure. We'll be planning generic, upgradable research and production facilities. But all of that starts with people who know how to think clearly about hard problems using the best tools available.
This series is Step 1.
"The distance between curiosity and discovery used to be measured in degrees, funding, and institutional access. AI reduced it to a methodology and a laptop. This series is the methodology."
