if you have interesting books/resources/ideas feel free to send me an email at satpathy(dot)abhishek001(at)gmail.com

1. Programming Massively Parallel Processors.
2. Active Learning for modeling expensive simulations in partnership with LMI.
3. Lots of practical automations at Forge ranging from email agents to logging alerts.

1. diff-sentry a stab at scalable code analysis for OSS security - link
2. a transcript timestamp annotator tool (meant for curating a dataset similar to Ego-Exo4D) - repository
3. a full pipeline built in Python to extract symptoms from natural language using Apache cTAKES and SciSpaCy and validate the results against pre-determined ground truth - repository.
4. a named entity recognition analysis eval pipeline - repository.

1. CS 3130 - Computer Systems and Organization II: Virtual memory, caches, pipelining, and some other low-level stuff including how attacks like Meltdown and Spectre work behind the hood.
2. CS 3100 - Data Structures and Algorithms II: An upper-level course on algorithms with an emphasis on problem-solving techniques with graphs, divide-and-conquer algorithms, greedy algorithms, dynamic programming, reductions, and some basic machine learning.
3. CS 4501 - Reinforcement Learning: An introduction from bandit algorithms to deep-Q-learning meant to teach you about the principles of Reinforcement Learning without getting too nitty-gritty with proofs and the mathematics behind the algorithms. We worked through chapters 1-13 of "Reinforcement learning : an introduction / Richard S. Sutton and Andrew G. Barto" skipping a few minor things along the way.
4. ECE 2410 - Machine Learning: An introduction to some of the principles of machine learning beginning with unsupervised vs. supervised learning and ending with hyperparameter optimization and neural networks.
5. CS 3120 - Theory of Computation: Finite State Automata, Circuits, Context-free languages, Turing machines, just the usual.
6. CS 4501 - Algorithmic Economics: How modern economics combines techniques in machine learning and statistics with economic knowledge to create efficient markets. Mostly an introduction where we learn about the basics of how reinforcement learning techniques, game theory, and tools like linear programming can be applied in an economic context.