Longer description of your proposed project

Flashbang School is a college-level learning platform built on short, fast-paced interactive videos, spaced repetition reviews, and simulation games. Course topics under consideration--selected to fill gaps in current educational media--are:

• world history of the present (1989-2022)

• cognitive security and critical thinking

• systems and strategic thinking

• philosophies of AGI, superintelligence, and cyborgism

• science/engineering of solarpunk, permaculture, and renewable energy

• collaborative governance

Each lesson is broken into modules of short (~3-5m), interactive videos, with timed/gamelike post-video reviews that assign relevant flashcards to the learner's deck for long-term retention.

At the end of each lesson, learners are tested with a simulation game in a role-playing environment, drawing on their lesson knowledge to progress.

Learners can at any time generate an up-to-the-day transcript (as a personalized web page on the site) which demonstrates their lesson performance and retention of the material over time and with high granularity.

I hope these ideas serve as the foundation for more experimental features as the platform grows. Ultimately, I aim to create in Flashbang a "laboratory school," in which many high-variance ideas about online education can be served to the public and tested/improved upon.

Nearterm Roadmap

Over the course of the year I intend to incrementally roll out a full lesson demo on the platform.

By late spring I'll enter a closed beta with a few testers and a rough prototype of a single, four module lesson with an exam (an exploration of eastern Europe in the years leading to the fall of communism for the modern history course.)

By the beginning of fall I intend to have a more polished demo lesson, with more of the school's infrastructure built on the platform (i.e., the 'transcript' feature and the first couple course's full syllabi) that I'll open up to a public beta.

Midterm Roadmap

• Designing research and content creation agents to speed up lesson creation

• Hiring SMEs to consult on and evaluate course content

• Completing the first two courses (modern history 1989-2000, cognitive security/critical thinking)

• Exploring accreditation or an informal accreditation-like system

Path to Revenue

• Freemium courses, where 'synthesis' modules/more games/more interpretive, supplemental and experimental content is paywalled

• Developing a second track of courses on more in-demand subjects for employment (software engineering/ML, for starters) and charging for courses

• Affiliate linking course readings and paywalled MOOCs

Describe why you think you're qualified to work on this

I have a bachelor's degree in philosophy, where my focus was on Deweyan pragmatism and educational theory, the principles of which I remain deeply committed to.

My belief that education is the "fundamental method of social progress and reform"(1) led me to pursue teaching and curriculum development as a career. I began teaching LSAT prep in college and went on to lead a mentoring team at an afterschool program in New Orleans for several years. I was also fortunate to work briefly at Synthesis School, an experimental K-8 startup that ran simulation games with students. At my current day job, I'm a curriculum developer at an online coding school specializing in machine learning.

In addition to instructional design expertise, as a machine learning specialist I'm familiar with the current best practices of AI engineering and use them both to make the platform more adaptive and to speed up production of content.

I first learned to code via spaced repetition self-study and taught myself machine learning using the same techniques. I later developed a system for auto-generating Anki flashcards with a GPT-4/embeddings agent and MOOC transcripts that I've since fallen in love with, processing a dozen+ MOOCs and Great Courses into a giant Anki deck. You can see a few of the courses I've Ankified on my website. I believe spaced repetition is still underappreciated, even by its fans, and especially when augmented with LLMs.

Pushing the boundaries of what spaced repetition can do is one of my core motivations, as it has had such a positive influence on my life.

1 - Dewey's "My Pedagogic Creed," Article 5 (a very good read!)

Other ways I can learn about you

flashbang's substack: flashbangapp.substack.com

portfolio site: joeholmes.dev

linkedin: https://www.linkedin.com/in/joe-holmes-285212240/

How much money do you need?

My main blocker is time and my main goal in the near and mid term is to secure a workable FT salary for Flashbang development. My second most important expense is hiring subject matter experts (SMEs) for consultation, to check for holes/flaws in my content and to help raise the bar for the courses.

Raising these funds now will provide a vital early boost to this mission. It will shorten my timelines for FT availability and help me experiment with SME contractors.

What I can accomplish with minimum funding

• Significantly speed up my timelines for savings/runway to transition to a part-time position, freeing up 40+hr/wk to work on Flashbang

• Explore hiring a history SME to collaborate on Flashbang's history syllabi/content

What I can accomplish with full funding

• All but guarantee I'll have the means to step into a part time role with an employer by January 2025, freeing up 40+hr/wk to work on Flashbang

• Explore hiring history/philosophy SMEs to collaborate on Flashbang's history and critical thinking syllabi/content

• Experiment with a few very small marketing campaigns to see what resonates with my target audience

Links to any supporting documents or information

My Substack devblog is at flashbangapp.substack.com.

The current v1 (unpolished hackathon version) of the site is at flashbang.school. It uses existing open courseware instead of my own assets and hasn't been maintained since the hackathon, but it might give you some idea of where I'm headed.

Estimate your probability of succeeding if you get the amount of money you asked for

In this context I'd define success as seeing > ~500+ users complete a finished course and develop a daily SRS review practice on the platform. I picture Flashbang becoming a boutique prototype where the limits of what's possible in edtech are explored. If it can be profitable enough as a business to permit me to work contract/PT while continuing to learn and teach its content, I'll be overjoyed.

I'd rate success according to this definition as somewhere near 70%, with funding. Higher education is broken enough, and there are enough genuinely novel ideas in my roadmap. I have the engineering chops to develop the platform solo, and so long as I have a small budget to consult phDs in the relevant fields I'm not concerned about the facticity/depth of the content. The biggest challenge I foresee is in marketing, which is why I intend to perform some small experiments on several channels with funding (likely educational YouTubers to start).