uploaded a small csv of some contacts and asked it to remind me of conversation status with them
CODE
-
Deep Research Web Feature: Select Sources and OAuth Integration
By
–
on the web, when you select deep research, it shows a new menu item where you can select sources, and you can just oauth into stuff like hubspot box gmail
-
RASP Programming Language Compiles Algorithms to Transformer Weights
By
–
another incredibly underrated paper: Thinking Like Transformers (Weiss et al, 2021) presents RASP: a programming language that compiles to transformer *weights*. can implement sort(), bincount(), etc. seems important. why don't interpretability people care about this?
-
Managing Scale: Avoiding Production Outages with Redirects
By
–
Lots of people using at scale, generally I agree we should not redirect, but trying to not cause production outages.
-
Reducing Preview Models for Faster Better Development
By
–
In the future, we hope to do fewer preview models and avoid some of the churn this causes folks as we fix bugs, keep hill climbing, and address developer feedback. Please keep the comments coming and we will keep pushing to ship faster + better + more clearly : )
-
Google Redirects Gemini 2.5 Pro Preview Endpoints to GA Model
By
–
Heads up: we will be redirecting "gemini-2.5-pro-05-06" and "gemini-2.5-pro-03-25" endpoints to our GA model gemini-2.5-pro on Jun 26 at 10 AM PST. The latter (03-25) currently points to the (05-06) model. Thanks for being on the 2.5 Pro journey with us, we learned a lot : )
-
Inverting Binary Trees: Interview Problem Not CS Concept
By
–
In fact, "invert a binary tree" does *not* exist as a computer science concept at all. It was created entirely as an interviewing problem. If you ask someone "can you invert a binary tree" then you're testing whether they know the definition of this interview-specific term.
-
Replit overlooked in AI talent acquisition race by Big Tech
By
–
I don’t hear Replit mentioned much in all this talk of AI talent and company acquisitions from Meta, Apple, OpenAI, Microsoft and others. Maybe that will change soon.
-
Top 50 LLM Interview Questions: Comprehensive ML Guide
By
–
Top 50 LLM Interview Questions! A comprehensive resource that covers traditional ML basics, model architectures, real-world case studies, and theoretical foundations.
-
Hugging Face Auto-selects Inference Providers for Models
By
–
or…. You can now use @huggingface Inference Providers with provider="auto", to automatically select a provider for your model. It will pick the first available provider based on your preferred order set in hf[.]co/settings/inference-providers