I wish I knew the havkernews equivalent for chemistry, finance, etc. The closest thing I've found was https://www.foundmyfitness.com/news for nutrition.
Outside of the academic literature, it is quite an insular field, so you have to work to find interesting ideas. If you can make money, you probably won’t be sharing much with anyone. A notable exception might be Rob Carver’s blog (https://qoppac.blogspot.com/).
From a CS background, a simple path to see if you might enjoy “quant investing” is to read some old papers from e.g. AQR about factor investing (Momentum, Value) and try to implement it with Interactive Brokers or similar. You should be able to roll something together with very little capital.
On the “quant trading” side of things, try the Avellaneda paper for statistical arbitrage, do some digging about RenTech, read the LTCM book, etc etc. Not something you can do without a lot of infrastructure behind you, but there’s enough information around to work out if it’s interesting.
I'll give the paths that I saw available to someone with a usual CS background. Not hard rules, but the distinction between developer (support) & trader (revenue generation) will be harder to overcome the later on in a career that it happens. Trading/research roles are typically compensated as a direct function of the money you generate, which is where the real upside is.
1) Quant Developer: Trading infrastructure, working on research platforms, taking algorithms created by researchers & deploying into production. Wouldn't want to do this at a bank, but at a hedge fund or proprietary trading firm it'd be fun. Compensation is good, lower variability, but somewhat more limited upside. DE Shaw, 2 sigma, Jane Street, etc.
2) Quant Researcher: A high quality quant masters degree at a minimum (think Baruch MFE), but often a PhD. Working on systematic trading research, market making algorithms, optimal hedging for large books of derivatives, etc. Usually a decent salary, but total compensation varies wildly based on performance.
3) Trader: More and more similar to #2 these days, but more focus on execution of strategies and real-time action rather than research, and more directly responsible for PnL. Optiver, IMC, SIG, etc. A quant masters degree probably isn't required to get a foot in the door, but I think it would help in the long term (e.g. helps build more rigorous understandings of why what you're doing might be working).