Intro topic: Getting an entry-level job

News/Links:

• Mario Kart 64 Fully Decompiled
  • https://gbatemp.net/threads/mario-kart-64-decompilation-project-reaches-100-completion.671104/
• Q-Learning is not yet scalable
  • https://seohong.me/blog/q-learning-is-not-yet-scalable/
• Grover’s Algorithm
  • https://www.youtube.com/watch?v=RQWpF2Gb-gU&vl=en
• OrangePi has a RISC-V SBC
  • https://linuxgizmos.com/orangepi-rv2-a-cost-effective-risc-v-board-with-m-2-2280-slot-and-dual-gigabit-ethernet/
    
    

Book of the Show

• Patrick
  • The Will of the Many (James Islington)
    • https://amzn.to/44Dznsz
• Jason
  • The Intelligence Trap
    • https://amzn.to/3TqoKCB
      
      

Patreon Plug https://www.patreon.com/programmingthrowdown?ty=h

Tool of the Show

• Patrick
  • Pokemon Odyssey
    • https://www.reddit.com/r/PokemonROMhacks/comments/1l9zdta/pok%C3%A9mon_odyssey_final_release/
• Jason
  • Netflix Games
    • https://play.google.com/store/apps/dev?id=6891422865930303475&hl=en_US
      
      

Topic:

• Why
  • Speed up development
  • Catch errors faster than type checking/compiling
  • Writing tedious boilerplate code
  • Ask questions and learn local information
  • Look good for hiring managers
• How
  • Extensions for VSCode & other IDEs for inline suggestions
  • Chat with a selection/file
  • Command-line Tools run at the root directory
  • Local vs Cloud
• Examples
  • Copilot (VSCode extension)
    • Use the experimental mode
  • Cursor (Custom IDE)
    • Jumps to suggest changes in other places
    • Similar to copilot experimental mode
  • RooCode (VSCode extension)

Intro topic: Video Game Prices

News/Links:

• Step one: Jump in the Lava - Abyssoft
  • https://youtu.be/WdadpHLAfdA?si=oXYnhB0EdkR_RaPE
• Scalable world models for continuous control
  • https://www.tdmpc2.com/
• Clever code is probably the worst code you could write - Engineer’s Codex
  • https://read.engineerscodex.com/p/clever-code-is-probably-the-worst
• A new, open source text-to-speech model called Dia has arrived to challenge ElevenLabs, OpenAI and more
  • https://venturebeat.com/ai/a-new-open-source-text-to-speech-model-called-dia-has-arrived-to-challenge-elevenlabs-openai-and-more/
    
    

Book of the Show

• Patrick
  • The Muscle Ladder - Jeff Nippard
    • https://amzn.to/44Dznsz
• Jason
  • Metaphysics of War
    • https://amzn.to/4jMjvZ5
      
      

Patreon Plug https://www.patreon.com/programmingthrowdown?ty=h

Tool of the Show

• Patrick
  • Pokemon Trading Card Game Pocket
• Jason
  • Phi-4
    • https://huggingface.co/spaces/microsoft/phi-4-multimodal
      
      

Topic: Memory Management

• Motivation
  • Avoid thrashing / crashes
  • Allocate resources efficiently
  • Keep high uptime
• Where
  • OS Level
    • Heap management
    • Virtual Memory
  • Language/Compiler Level
    • Cpp
    • Garbage collection
    • Ownership
• Tools
  • Instrumentation
    • Export to Datadog / Grafana
  • Python: psutil & tracemalloc
  • Valgrind
• What to do when your program uses too much memory?
  • Reduce data sizes
    • Compression
    • References
    • Lazy initializer
    • Generators & Back Pressure
  • Ring buffers
  • Arena allocators
  • Disk based caching

Intro topic: Grills

News/Links:

• You can’t call yourself a senior until you’ve worked on a legacy project
  • https://www.infobip.com/developers/blog/seniors-working-on-a-legacy-project
• Recraft might be the most powerful AI image platform I’ve ever used — here’s why
  • https://www.tomsguide.com/ai/ai-image-video/recraft-might-be-the-most-powerful-ai-image-platform-ive-ever-used-heres-why
• NASA has a list of 10 rules for software development
  • https://www.cs.otago.ac.nz/cosc345/resources/nasa-10-rules.htm
• AMD Radeon RX 9070 XT performance estimates leaked: 42% to 66% faster than Radeon RX 7900 GRE
  • https://www.tomshardware.com/tech-industry/amd-estimates-of-radeon-rx-9070-xt-performance-leaked-42-percent-66-percent-faster-than-radeon-rx-7900-gre

Book of the Show

• Patrick:
  • The Player of Games (Ian M Banks)
    • https://a.co/d/1ZpUhGl (non-affiliate)
• Jason:
  • Basic Roleplaying Universal Game Engine
    • https://amzn.to/3ES4p5i
      
      

Patreon Plug https://www.patreon.com/programmingthrowdown?ty=h

Tool of the Show

• Patrick:
  • Pokemon Sword and Shield
• Jason:
  • Features and Labels ( https://fal.ai )

Topic: Reinforcement Learning

• Three types of AI
  • Supervised Learning
  • Unsupervised Learning
  • Reinforcement Learning
• Online vs Offline RL
• Optimization algorithms
  • Value optimization
    • SARSA
    • Q-Learning
  • Policy optimization
    • Policy Gradients
    • Actor-Critic
    • Proximal Policy Optimization
• Value vs Policy Optimization
  • Value optimization is more intuitive (Value loss)
  • Policy optimization is less intuitive at first (policy gradients)
  • Converting values to policies in deep learning is difficult
• Imitation Learning
  • Supervised policy learning
  • Often used to bootstrap reinforcement learning
• Policy Evaluation
  • Propensity scoring versus model-based
• Challenges to training RL model
  • Two optimization loops
    • Collecting feedback vs updating the model
  • Difficult optimization target
    • Policy evaluation
• RLHF & GRPO