Yes! It basically picks up where the LLM from Scratch left of. I.e., we start with a pre-training LLM and then implement and add reasoning techniques.
(However, it can also be read as a standalone book)
LLMS
-
New Book Builds on LLM from Scratch with Reasoning Techniques
By
–
-
Learning Bits Through RL and SFT: Research Insights
By
–
best paper or blog i've read in a while, highly recommend! John is brilliant and his research sets an example for the rest of us. recently i too have been thinking deeply about how many bits might be learned via one step of RL or SFT.. if you're thinking about this too, lmk!
-
Improving Chat Completions API conversation handling and deduplication
By
–
That's the idea, yeah – that and do a better job handling Chat Completions API style things where a full conversation is passed at once (I want to dedupe that into an existing tree)
-
LLMs as Code Writing Assistance: Measuring Development Productivity
By
–
I think it's useful because it's a direct measure of how much literal code I would have needed to write myself if I didn't have help from the LLMs
-
PromptCoT 2.0: Scaling Prompt Synthesis for LLM Reasoning
By
–
PromptCoT 2.0 Scaling Prompt Synthesis for Large Language Model Reasoning
-
LLMs as Custom Solution Builders Replacing External Dependencies
By
–
It's so different Adding an external library as a dependency is a big overhead: you have to evaluate that it does what you need, and does it well, and tha it's liable to be maintained by a responsible team long into the future LLMs can build you the exact solution you need
-
Developers Excitement New AI Model Releases Claude 4.5
By
–
People generally get curious during launch of new phone models, cars, web series. Whereas Devs gets excited when new AI models get released Everyone is talking about Claude 4.5 Sonnet
-
LLM Generated Code as Custom Programming Solutions
By
–
Because the LLM generated code is custom code written to-order to solve a specific problem Similar to if you hired other programmers and gave them a detailed spec for the code you needed them to write
-
Claude Sonnet 4.5 Now Available for Perplexity Pro Subscribers
By
–
Claude Sonnet 4.5 and 4.5 Thinking are now available for Perplexity Pro and Max subscribers.
-
AI Models Now Enable Researchers to Replicate Economic Studies Independently
By
–
Models are good enough now to start significantly speeding up researchers across fields. Sonnet 4.5 using the code execution and file creation feature in claude dot ai was able to replicate all the experiments in a published econ paper just from the raw data.