Business update, week 8

September 14, 2025 at 6:47 PM

Hi everyone!

Apologies for missing the update last week. I was moving apartments last weekend and decided I was best off focusing on unpacking at the new place so I could get back to work as soon as possible. I’m all moved in now and was back at it this week.

Over the past couple weeks I’ve been working on two things:

1. Gauging interest in a ‘Forecasting 101’ guide where I describe the strategies I used to win several forecasting tournaments over the last few years
2. Pivoting to a ‘robotics job tutor’ idea which I intend to spend the next month on.

Forecasting 101 Update

At the time of my last update, I had just published a pre-order page for a Forecasting 101 guide and started posting educational forecasting-related things on LinkedIn, with links to the guide at the end. One person pre-ordered the guide shortly after my first LinkedIn post, and my plan was to continue making posts for a week to see if I could get 5-10 complete strangers to sign up.

I did eventually get a 2nd person to sign up, but nobody else after that. Interestingly, I was able to get quite a bit of engagement on my LinkedIn posts - one in particular reached almost 40,000 users! But those 40,000 views only translated to around 40 visits to my pre-order page and no purchases.

I suspect that as this post got more engagement, the algorithm was mostly showing it to people who were reading it because of the intriguing intro, but who had no real interest in a how-to-forecast guide.

My overall takeaway from this product idea is that, while it would be appealing for a very specific type of person, this particular audience is (a) a bit too niche, and (b) has low willingness to pay. I have (relatively) good access to the forecasting community, but we’re still only talking about maybe a few thousand people, and of that group, most of the people probably treat it as a low-commitment hobby, not something they’d invest money into. I’m not really solving a problem that’s urgent or painful.

I also struggled to figure out what a more scalable business in the forecasting space might look like. In theory I could sell some kind of consulting service to industries where accurate geopolitical forecasting is important (I believe this is one way Good Judgment Project and Metaculus make money), but I couldn’t come up with any clear path to getting started with that type of thing.

In the process of posting and networking in the forecasting space, I connected with the cofounders of an AI-based forecasting company called Mantic. Mantic was actually a participant in some AI vs. Human forecasting projects I did with Metaculus last year and they recently announced their startup publicly.

I had a call with one of Mantic’s co-founders earlier this week to chat about what they’re building and general forecasting stuff, and the other co-founder was one of the two people who pre-ordered the forecasting guide.

When I let them know that I wasn’t planning to finish the guide, they responded and said they’d be willing to fund it themselves (presumably they wanted to use it to help design/train their forecasting models). I’m working out the details, but it’s possible I’ll still end up writing a forecasting guide in one form or another and get paid a non-trivial amount for doing it.

Robotics ‘Job Tutor’ Idea

I decided to move on to a different space that I have expertise in, and which has a much larger pool of potential customers - robotics.

The idea I’m working on now is a service where I help early-career software engineers get jobs at robotics companies. The basic idea looks something like this:

• I review your resume and give tips on what to include / what to avoid
• I give some general guidance on how to approach the job hunt - I’ll provide a checklist with things to do each day, each week, and each month
• I help you find jobs to apply to and optimize your resume for each application
• I give practice interviews on-demand once you’ve got an interview scheduled - I can tailor these to the specific roles you’re interviewing for
• I give weekly ‘homework’ about different robotics-related topics to help you stay sharp on different topics
• I help you with monthly ‘portfolio projects’ - bigger projects that you can add to a portfolio that showcase and develop your robotics skills

The service would have a high cost (probably $1000+) and I’d offer a guarantee where if you follow my guidance and don’t get a job offer within a set timeframe (maybe 4 months), I refund the money.

I’m currently working on a ‘lead magnet’ - a valuable, free resource that will be useful to people in my target audience, which then mentions and links to my paid offering. I’m building a ‘how to get a job in robotics’ guide that gives a comprehensive overview of all of the skills, knowledge, and experience an applicant should have to stand out. It covers things like:

• Core software engineering skills every robotics software engineer needs
• Common robot software architectures / system design
• Major subdomains within ‘robot software’ - motion planning, state estimation, control systems, simulation, etc. Covers key concepts, algorithms, and use cases
• Practice exercises and sample interview questions for every topic covered
• General advice on how to approach the job application process

Because the guide is free and useful on its own, I can post it in places like Reddit without having my posts removed for self-promotion / advertising. And I also have plenty of other options for attracting clients - I could make YouTube videos where I explain different robotics concepts, I can post on LinkedIn, I can even do direct outreach to people who are actively job seeking in robotics.

There’s a lot that I like about this idea:

• As with forecasting, it’s a topic I have plenty of direct experience in. I’ve personally had 3 robotics / robotics-adjacent jobs, and have been both an applicant and an interviewer at each company.
• I’m solving a problem that is real and urgent for my target customer. Job hunting is stressful and early in your career, it’s hard to know what to focus on and what you might be doing wrong.
• This is a problem that I’ve dealt with personally. I would have paid for this kind of service if it existed when I was going through previous job application cycles.
• The money-back guarantee makes the offer compelling - it makes sure my incentives are aligned (I only get paid if the client succeeds) and de-risks it for the client (they only pay if they land a job).
• The guide I’m building (and any YouTube videos if I go that route) will be a great resource and something I’m proud to put out in the world, even if it doesn’t end up attracting any paying customers. Worst case scenario I’ve created something cool and learned a bunch in the process.
• It’s a high-ticket, low volume business model. I only need a few clients per month for it to be a sustainable business.

I’ve made good progress on the ‘getting hired guide’ and it will be my main focus for the rest of the month. Right now I’m writing a rough draft of all the content, and once I’m done with that, I’ll turn it into some kind of interactive website.

Once the guide is live, I’ll switch my focus to recruiting a first cohort of clients to bring through the program. The first few clients will be heavily discounted or free - my goal with the first batch will be to figure out how to make my service effective and hopefully get some strong testimonials to drive future sales.

Lots of work to do for this idea but I think it’s very promising! More updates to come.

What are my goals for next week

Main goal: complete half of the rough draft of the ‘Fundamental Robotics Concepts’ section of my robotics guide. There are 11 sub-sections, so this means finishing at least 5.5 of those. This section will constitute the majority of the overall guide.

I haven’t been doing a great job of hitting my goals lately… this week’s goal is very achievable, it just requires me sitting down and doing focused work consistently all week. So I have no excuses if I don’t complete this one!

What’s working / what isn’t

The last two weeks have been a bit weird because of the move, so I’ll hold off on drawing any major conclusions. I lost about a week of work all things considered but now am back into a normal work schedule.

The break was helpful psychologically, I was a bit stressed about the move and maybe a bit overworked in the lead up to it. My early impression is that the new apartment is a better work environment than the old one, mainly because it is much brighter and my work desk isn’t right next to my bed.

So long as I stick to my goal this week and work on the robotics guide all week, I think I’ll do some kind of ‘lessons learned after two months’ update next week, rather than a usual work update.

That’s all for this week!

Jake