Keep on to blur preview images; turn off to show them clearly
just putting this out there again: “models are non-fungible in their harness” we’re all trying to build great AI systems and products, that means all the components need to work well together, it’s more than just dropping in a model the new results coming out are real, same model + diff harness = way different results harness engineering is in!
building agents and harnesses, prev @awscloud, phd cs @ temple
RT @relace_ai: Fast agentic search is here! Try it out for yourself on our model playground https://t.co/6ypeg1iN8a
Cartoonist, Engineer, PM, Partner @a16z investing in infra & AI Prev Product lead @HashiCorp, Founding Eng/PM @Transposit. Eng @AppDynamics. Opinions = own.
If my toddler is any indication, @Waymo will be the default way for all of us to take naps on the go
Partner @A16Z - started @speedrun, prev @Initialized, 2X founder - acquired into @Farmville @Zynga
Looking around the timeline, way too many people on here, including folks that have been around for a while, have become so cynical and so randomly bitter day in and day out. Why?
Asst professor @MIT EECS & CSAIL (@nlp_mit). Author of https://t.co/VgyLxl0oa1 and https://t.co/ZZaSzaRaZ7 (@DSPyOSS). Prev: CS PhD @StanfordNLP. Research @Databricks.
Making LLMs Bayesian would solve all their problems and usher in the age of AGI. And all you have to do is run the model a million times for each question and average the results. (Disclosure: Nvidia paid me to say this.)
Professor of computer science at UW and author of '2040' and 'The Master Algorithm'. Into machine learning, AI, and anything that makes me curious.
RT @thejustinwelsh: So many people ask "How do I start?" for longer than it would've taken to just figure it out.
The $10M Solopreneur | Helping 100,000+ experts turn their expertise into income.