The difference here is that most investors in these kinds of "securities" being issued by tech companies would demand more data than the investors in securitized mortgage assets did. In lieu of performance data, investors in MBS relied on ratings from ratings agencies that were dependent on the business of the banks that issued the securities. The failure of ratings agencies to accurately assess the risk in the securities is the reason the MBS vehicles grew in popularity (they all looked AAA!) and also why they exploded.
Yeah, I agree: it's super straightforward. It's hard for me to understand how people could find it inscrutable / confusing. Pandas is a LIFESAVER for anyone who needs to manipulate datasets programmatically.
Big fan of pandas as well... but I have to admit there are some things at the beginning were like wtf... some of which are now deprecated thankfully. But more on topic to this post I really can’t see why someone would want to solve speed on small datasets by incorporating numpy into a new form of pandas... both projects are so established, why would you attach your project to some other dev who know has to keep pace with numpy and pandas improvements when you could just import pandas, import numpy and be done with it?
- Get your accounting in order. Pay a professional to do it; don't try to do it yourself.
- Take care of yourself. You'll probably be flying a lot; don't take the cheapest routes, tell the client it's business at >4 hours in the air, etc.
- I unfortunately get brought into a lot of situations where a CEO / VP is looking for justification in firing someone or making a big structural change. In these cases, it's important to keep in mind that you're working for the person who is paying your fee and to not get hung up on trying to fix the situation. It sucks.
- Charge a lot. Think up the largest number you believe would be reasonable for your services and multiply that by 1.5x. Consultants are supposed to be expensive.
- You'll be amazed by how many people reach out and want to take phone calls to go on fishing expeditions / mine you for free value. The way I deal with this is: A) I ask very pointed questions about what, exactly, they're looking for in an engagement via email before I take a call. If they don't have a compelling need, I decline ("I don't think this is a good fit for me." -> be blunt) and B) I have developed a 6th sense for whether a company can afford me / the project sounds viable. If you have enough inbound, don't be afraid of false positives in turning down phone calls.
- Don't bill hourly. I charge per project on the basis of value; if a potential client fights you on this, reject them.
- Try to get work on a retainer basis. You're being paid to not only do work but to be available -- you're a service provider. This is nice because the revenue "stacks" up and you also get more integrated into the team this way as it frames you as a source of insight / wisdom. If you're using consulting as an entry point into a new role, this is very effective.
- Build something proprietary: an Excel model, some tech, an audit framework, a taxonomy, whatever. Brand it. That's now "your thing" and you can set a market price for it that is divorced from incremental work.
- Build out a proprietary funnel for business: a blog, Twitter presence, whatever. If you have to go find business on your own / do sales, you can't charge as much as when people come to you.
- In general, I think it's good to be skeptical and to err on the side of telling people "no." There are a ton of lightweights / window shoppers / sleazy people out there; get good at detecting them and rejecting them quickly.
- NOTHING IS FREE. You don't pitch, you don't do exploratory onsites, you don't give samples, you don't take "no-agenda meetings," you don't let people "pick your brain." You are a consultant and if people want your insight, they need to pay you for it. Internalize this phrase: F* YOU, PAY ME.
