- Experiment as much as possible - Keep your feedback loop as short as possible between reading and doing - Visualize things (debug outputs, scratchpad

- Experiment as much as possible - Keep your feedback loop as short as possible between reading and doing - Visualize things (debug outputs, scratchpad of short lived diagrams, etc.) - When possible, envision what the finish line looks like and go backwards from there For the first one, if you think solution A, B or C might be good then don't spend a huge amount of time theory crafting or wondering "what if" while you torture yourself to pick the "right" one for your use case. Do your basic research and make a proof of concept with all 3 and pick the one that feels best after you have experience with all of them. Often times this isn't much slower than going down a deep theory craft rabbit hole and the result is going to likely be better and you'll feel better about it in the end. The last bullet is especially useful for building tools or APIs where you can start with using your tool / API as if it does exist exactly how you want it in a best case scenario and work backwards to implement the solution you want. Break down the steps along the way.