# New Daily Plan --- # Goals #### Big long term goals **Continue building skill set and experience to allow for freedom, autonomy and optionality** * Potential ideas: * Consultant (I could do everything. Set up your data pipeline, database, perform the actual analysis, and soon. Target would be small companies. Would combine so many different skill sets. The one your are building now is hardcore engineering and product skills) * Lead R&D engineer, lead data scientist at small startups #### Mid level (rest of 2021 goals) **Get promoted to Senior AI Engineer by years end** * Impress Tom. Michael will likely be talking to Tom (to a degree) to determine if I should be moved up to a Senior AI Engineer. The more I can impress Tom, the better. * Push hard to accomplish our goal user outcomes: * **Facilitate Understanding:** I need to be a subject matter expert in the data and tooling available to me in order to deliver domain-relevant insights quickly * **Deeply Understand Relationships in Data** * **Ensure Accuracy:** I need to ensure that the insights I deliver to business stakeholders are accurate * **Validate Raw Data** **Gain a deep understanding of the bridge between the discrete and the continuous**: * This includes: * discrete differential geometry (algebraic topology as well in this course) * ML on graphs * graph algorithms and DS **Big R&D advance for AI** * Build upon attribute conditional dependency analysis * How does this relate to Toms Attribute Space work? **Grow mathematical maturity** * so that you have a foundation that continues to be able to learn things more quickly * This is one of those things that is hard to really improve, but you clearly have made significant progress. It takes time and won't just happen, but it is *so* worth it once you have #### Tasks to help achieve my goals * Embrace all challenges on AI team. Anything that will help move them forward will likely be the most beneficial for me. * Introduction to higher math videos (60 hours. this will never be *top priority*, but it really should be done at some point. Never will be a *great time*.) * Topology * functional analysis * tensors * manifolds * differential geometry course CMU * Graph embeddings course * Algorithms & DS (probabilistic data structures) * Graph theory algorithms * Weak spot: optimization techniques (currently we don't do any of this at unsupervised, but review gradient descent, linear programming, neural nets, etc, would be useful. Not a big time sink, but good to refresh on) * New blog format (with dates so that I can see how much I have written each year) * 3 blog posts by end of year #### Next 6 week block * TODO # New Habits/Practices Going Forward #### Morning Ritual/Mantra A sheet with the following: 1. Rules & Constraints (what eats up your time, what won't you do, diet, exercise). this sets the constraints in which the game is to be played. 2. Gratitude (Visualize on idea that you may die at any moment today. Accept the notion that today may be your last day on this earth. It is amazing to simply be alive) 3. Long-term goals (5 years, ambitious, big goals - TODO: these need to be more concrete) 4. Short-term goals (2021 goals - just out of reach, but attainable with hard work, this should get you amped up!) 5. Visualize the day (what do you need to get done, how will you walk through the day) 6. Core principles that I strive to live up to (what are my core principles? Compounding, consistency, reliable, hard work, empathy, compassion, love, character, integrity, strength - physical and mental) 7. Values: 1. Be someone who ships 2. Alignment with goals #### What to do with interesting ideas during deep work When ideas pop into your head, write down and put aside, address later on. Ideas will keep coming, but you really want to focus on the task. Only interruptions allowed are water, coffee, bathroom. #### Each Morning... * Open your daily note/plan in obsidian * It will contain * morning ritual * if you have downtime tomorrow you should focus on _____? #### Deep work * I should start prioritizing unsupervised problems for my deep work. In general, as long as the problems are interesting, this should be prioritized as your deep work. * As crazy as it sounds, complex math in some ways is becoming *easier* to learn (via following a video, book, etc) than creative and complex tasks for unsupervised. #### Organization techniques * Create project for rest of the year (Epic) * then, 6 week block project gets pulled from * create an asana project for the 6 week block * Create a new TODAY project? * Create a math and DS board that is AI/math/tech back log * Need a better way to organize and keep new content in order * Plan an hour long block weekly to reorganize things in this box * Plan another hour long block for weekly planning * track time spent working on 6 week cycle (via a specific project in toggl) * sync with the asana project that you want to work on * so maybe you can tag it? or call the project deep work block 1? #### Schedule * will have 2 hours on a wednesday night * 3 deep work sessions per day * 645-845 * 9-12 * 1-430 #### Planning * 6 week block that is aligned with * big picture goal * unsupervised personal goal * midterm goals, etc * And inline with how many hours you have! * 1 week block (after 6 week block) of: * reflection * planning new block * blog post writing * unsupervised ideas * synthesis * organize notes * #### Day to day practices to follow * Try not to rabbit hole too badly * Do less. Do better. Know why * What would make today great? * At the start of day orient on big picture goals (5 years), rest of 2021, 6 week block. * Figure out a good cadence for planning week. How much time for unsupervised problems, for learning totally new things, etc. * Have your mornings planned at the start of the week * Anki (2 obsidian notes per day essentially, this will help ensure that I frequently review my notes and that I take good notes) #### New tooling approaches * obsidian as a base. New ideas should go in to obsidian incubation (TODO: figure out how to get obsidian on my iphone) * screenshot presentations to save time, obsidian as a base, link drawings from notability... # Ideas to Consider * [Deep work index card deck](https://www.calnewport.com/blog/2015/01/18/deep-habits-use-index-cards-to-accelerate-important-projects/)