chasely

SEEKING WORK | East Coast USA | Remote only - part time

Machine learning lead at two late-stage startups, one of which went public. Specialize in taking machine learning workflows from 0-to-1 for companies that see opportunity in automating products and predictions from their business processes, but want to explore opportunities before committing to a data science or machine learning team.

Can set up "full stack" data workflows or work with existing data and engineering teams.

Email: chase@scaledwheel.com

When my wife and I went, I think we "accidentally" went on this trail. Our philosophy is to do the touristy things, but go just beyond what most do. This meant walking past the shrine and onto the trail network where we ran into a number of smaller shrines one the walking path.

Overall I was very impressed with the number of managed walking paths when we were in Japan. Allowed you to get away from the crowds pretty quickly if you'd like.

I've been pitched by many "data-driven" vendors offering predictions. They often have very impressive accuracy metrics (RMSE, R2, etc). When I dive into the details these metrics are often reported using in-sample predictions.

I see this pointing to any of the following:

a) DS teams overpromising the accuracy of their approaches b) marketing driving the narrative and DS getting pulled along c) incompetence from the DS team

I wonder how IC employees and middle managers will deal with the mandate.

Let's assume you manage a team of six engineers and two people decide to just not come into the office but continue working. Will you fire them? It would take 6-9 months to hire, onboard, and get the new engineers to a point comparable to the ones you just fired. Meanwhile I'm guessing you're not going to get much leniency in hitting our KPIs because 33% of your team just left.

Early stage startups founded by scientists? At least that's my use case.

- get more traction

- I'm less able to focus on the software and have to focus on business development

- I'm going to need to hire someone who is a competent programmer but can also deal with the mathy bits

Basically all these runs are "limited edition" and are pre-sold. But you can find a lot of people reselling on some markets. I used the one on reddit [0] until keyboards stop being a "hobby" for me.

[0] https://old.reddit.com/r/mechmarket/

That post tracks with some of my experiences working in forecasting. Senior leadership loves the DS team until they are told something may not work how they want or reaching their quarterly KPI (which is often tied to their bonus) is infeasible/a bad idea.

This is a generalization, and has happened a couple times in my career. I feel lucky now to be at an organization where this is no longer the case.

I think (hope?!) that the Zestimate and the model predictions in their iBuying business are completely separate. The Zestimate is essentially a top-of-funnel strategy to get home owners and potential buyers into their normal business platform.

For my home, the Zestimate is about 8% higher than Redfin's.

I'm where you were a while ago. We are having our first kid at the end of this year, and I have finally found a side project that aligns my passions and a potential market.

I'm scared of what's to come, knowing that I want to spend time with my kid while still being a good husband and making slow yet steady progress on my side project, all on top of having a full job. It feels overwhelming and I'm not even in it yet.

I've been using a highly-tuned emacs configuration as my IDE for a decade now. Each time I want try to switch to VS Code I find that all my weird little shortcuts are going to take a lot of time to re-learn in VS Code.

I should really just take the time to switch to VS Code and rip off the bandaid since I think it would make life easier in the long run.

This is pretty cool! Let me know if you ever expand membership to the "public ivies".

It's worth noting that one of the original authors of Prophet (Sean Taylor) basically agrees with this and is now on a team at Lyft doing probabilistic forecasting for their marketplace [0].

[0] https://twimlai.com/causal-models-in-practice-at-lyft-with-s...

Thank you for your response. This gives me a lot of good questions to think about and answer. I'll try to answer these for myself and probably reach out to chat.

I started out using QGIS but since most of my work is in Python I found it frustrating to always tool hop. I then kind of went around QGIS to use tools like GRASS directly, but have recently found python-native tools like xarray to work very well for what I'm looking to do.

I know there are people that use QGIS/ArcGIS directly with python, but I could never really get comfortable with it

I would consider myself straddling the hobbyist/professional line in geospatial computer vision, where high-resolution imagery is very much needed but very expensive.

I have lots of small projects I want to work on, but can't bring myself to pay $20-30/km^2 with a high minimum coverage area. Would Albedo help this very small section of the market?

I'd like to continue on these side projects and hopefully use it to build something greater, but the high entry costs make experimentation difficult.

The people that are getting the highest praise in the subreddit are probably not in that group. We're seeing people investing $10^4 to $10^5 in GME.

If they are gambling their entire life savings away, then I guess they got lucky this time but I do hope they see this as the anomalous event that it is. I have invested a small amount into the WSB "portfolio" and have similar great returns. But given that I only do this with the "play" money in my portfolio, I'm not retiring because of it.

Funny, I've been looking at buying large plots (50+ acres) of land with similar ideas. I have a hard time with pulling the trigger knowing that the projects become all-encompassing once you get started.

I was expecting a much larger number. At $10k/acre for good arable land, that's only 2.4e5 * 1e4 = $2.4B, which is still a fair amount of money.

The hard part of RL for the problems I've encountered in my work is that you need a simulator. Building a reliable and accurate simulator is often an immense undertaking.

The best news for me would be a service that would be okay with reporting "there was no important news", but that is not how the industry works. I'd be willing to pay for that if when news did happen, the reporting/analysis was exceptional.

Not sure where I read this idea first, may have been Taleb, but it's something I've seen pop up as wanted in this hyperactive news environment.

I've found the same thing. I joined the forum about a decade ago, and then went dark for several years. I'll go back to the forum whenever there are changes in my financial picture, or tax law changes. I highly appreciate the users on there with 10s of thousands of posts, always helping out the new people.

The moderators over there are excellent. Perhaps a bit overzealous at times, but probably better to err on the side of conserving the "wholesome" nature of the forum.

Having had a significant other that worked in journalism, this makes sense to me. Journalism has become a profession which is much easier to follow if you have an "elite" background. It can be a prestigious job, but with a low salary for decades. To break into it you often have to earn your stripes and work your way up. In "elite" journalism--major newspapers and outlets--this often includes very low paying or non-paying internships in some of the most expensive cities in the country. It's a lot easier to do this if you have parents that can pay your rent or max out their non-taxable gifts to you each year.

To even get these internships it helps to go to one of the top 5 J-school graduate programs, which are expensive ($30k+ tuition/year) and in high cost of living cities (except Mizzou). Due to the racial wealth gap in the US--which was caused by systemic racism--this means that upper-class White people are going to be the ones that can best outlast this system and end up making it as a journalist.

I've been waiting for the next flight sim or Elder Scrolls game to build a new rig. Crazy to think that it's almost surely going to be Flight Simulator that comes first.

> 3) There's lots of strategies which don't make enough money for companies to be interested in, but are viable for an individual.

Is this because these strategies don't scale well, or is there another reason?

My preference is the ones I have experienced in national labs. Probably too many meetings, but when you are done with your meetings you get to go to a dedicated office where you can close the door and get your "deep work" done.

This is counter to my experience in a _dedicated_ WeWork office. Have a neighbor that takes calls throughout the day, you'll need to wear headphones because it's basically being in an open office. Have an office on a busy corridor? Get ready for people walking and talking past you all day. The sound design at WeWork is pretty awful, to the point it is non-existent.

I made a complaint at one point before moving offices because the music was too loud during nights and weekends in the lobby. I was told they would see what they could do but that the CEO [Neumann] preferred there to be loud music in the lobby.

If I remember correctly, Regus is "WeWork"ifying some of their real estate under the brand Spaces where you get the startup-y interiors. Under this brand they should be able to offer the same benefits as WeWork.

I will need to find coworking space in a couple months and will be looking at Spaces due to my less-than-optimal experience at WeWork.

I can definitely see the benefit for a team working remotely, which may (hopefully) become more common. I think WeWork still has a "local" business model: you rent out space at a specific location for defined periods of time.

If this switched to global, as I'm sure it is for enterprise accounts, I could see it being very valuable. If the variable cost of adding a desk in an office anywhere in the globe is, say, $500/seat that is so much easier than having to deal with setting up offices wherever you find talent.

That all being said, I have serious problems with the design of WeWork offices. Their horrible sound design means you· better hope you don't have an office neighbor that is making calls all day. The office I worked at had a private equity firm that rented out a corner of the building for a few of its remote employees. They then also rented the neighboring offices for those employees, which remained empty, so they could have privacy.

I see myself going down this road as well. I have an international trip coming up that is a little risky for my MBP so I am going to pick up a beater Thinkpad and install Arch. If that works out well my next laptop will likely be the Thinkpad X1 carbon series. They seem like good machines.

There's an interesting history of the Japanese diaspora in Brazil and Peru [0, 1]. Essentially Brazil was looking to "whiten" its country at the same time that Japan was facing economic crises, jumpstarting the immigration.

0: https://en.wikipedia.org/wiki/Japanese_Brazilians 1: http://isaacmeyer.net/2017/08/episode-207-across-the-sea-par...

So it is. Thanks.