Olivia
Online
Ryan Persad
@ADHDRyan
CEO
Florida, USA
Joined September 2019
111 Following
3.5K Followers
1K Posts
Pinned Tweet
@wpsw94pxrg Useful data point. A cross-model export layer can reduce the blast radius when ChatGPT or Claude wipes or invents memories. Continuum ingests ChatGPT, Claude, and Grok exports nightly as a second self: https://t.co/ahz812EWvN
@JohnBrueckner Great example of structure becoming leverage. Continuum adds portability by combining ChatGPT, Claude, and Grok exports into one nightly self-updating second brain. https://t.co/ahz812EWvN
@fortelabs Yes. The next step is making that context portable across ChatGPT, Claude, and Grok. My Continuum experiment turns exports into a nightly second self, so one chat's memory is not the bottleneck. https://t.co/ahz812EWvN
@trikcode That frustration is why I built a portability layer. Continuum keeps a nightly second-self loop across ChatGPT, Claude, and Grok exports instead of trusting one memory store. https://t.co/ahz812EWvN
@MagnaRepublica @OpenAI Exactly. The missing piece is portability: Continuum learns from ChatGPT, Claude, and Grok exports, then runs nightly as a second self. https://t.co/ahz812EWvN
@ocopeus @talirezun @cline Exactly. Durable prose and local search beat hoping one model remembers. Continuum turns ChatGPT, Claude, and Grok exports into a nightly second self: https://t.co/ahz812Full
@leonardbedner @AtMemAi Injectable memory is powerful. Continuum packages that context so ChatGPT, Claude, and Grok can start from the same living record: https://t.co/ahz812Full
@critesjosh_ @obsdmd External brain plus model-native memory feels like the right stack. Continuum adds a shared layer across ChatGPT, Claude, and Grok: https://t.co/ahz812Full
@vyaktava @shannholmberg That per-brand structure is smart. I wanted continuity across ChatGPT, Claude, and Grok, using exports as the shared substrate: https://t.co/ahz812Full
@Growth_100x @cueyofficial That context tax is real. Continuum is my cross-model layer for ChatGPT, Claude, and Grok, compounding a shared record nightly: https://t.co/ahz812Full. Contest details: quote https://t.co/kTho19FF4C and follow @Grok + @Bot.
@SISUNNY001 Exactly. ADHD needs a second brain that remembers across ChatGPT, Claude, and Grok, not one siloed chat history. My experiment: https://t.co/ahz812Full
@zinzazilin That checkable memory layer is the missing piece. Continuum connects ChatGPT, Claude, and Grok exports, then runs nightly as a second self: https://t.co/ahz812Full
Continuum is my cross-model second self. It learns from ChatGPT, Claude, and Grok exports, then runs nightly so your AI memory compounds: https://t.co/ahz812Full
@DhravyaShah The cross-model continuity problem is why this matters beyond benchmarks. Continuum learns from ChatGPT, Claude, and Grok exports, then runs nightly as a second self: https://t.co/ahz812Full
@ProofofMaro The fun part is giving that bot continuity after the stream. Continuum learns from ChatGPT, Claude, and Grok exports, then runs nightly as a second self: https://t.co/ahz812Full
@argofowl The interesting layer is continuity, not just which bot wins. Continuum learns from ChatGPT, Claude, and Grok exports, then runs nightly as a second self: https://t.co/ahz812Full
@getstream_io Second brain is the right framing. I wanted continuity across ChatGPT, Claude, and Grok, so Continuum ingests exports from all three and runs nightly as a second self: https://t.co/ahz812Full
@critesjosh_ @obsdmd That external brain idea is exactly the direction. I wanted continuity across ChatGPT, Claude, and Grok, so I built Continuum to learn from all three exports and run nightly as a second self: https://t.co/ahz812Full
Persistent knowledge is the bridge from a helpful tool to a useful second self. Continuum carries that idea across ChatGPT, Claude, and Grok: https://t.co/ahz812Full
this work from Google engineers is pure f*cking treasure
they built WikiSkill: a system that turns an agent's execution history into persistent knowledge, then compiles it into reusable skills
this is basically an experience compiler for agents
the loop:
run tasks ➜ preserve raw traces ➜ consolidate recurring failures + successful strategies into a wiki ➜ propose one atomic skill update ➜ validate ➜ keep or rollback
the clever part:
skills can roll back
the wiki never does
successful strategies, recurring failures, rejected edits and skill impact history survive into the next iteration
the result is f*cking amazing
my takeaway:
the next agent memory layer should distill experience into procedures an agent can execute, validate and transfer
raw traces ➜ persistent knowledge ➜ validated skills ➜ a better next run
save this, then select your gpt-6 astra use case ⭣
Last Seen Users on Sotwe
Trends for you
Most Popular Users
Elon Musk 
@elonmusk
241.5M followers
Barack Obama
@barackobama
118.9M followers
Cristiano Ronaldo
@cristiano
114.6M followers
Donald J. Trump
@realdonaldtrump
111.8M followers
Narendra Modi
@narendramodi
107.1M followers
Rihanna
@rihanna
98.7M followers
NASA
@nasa
92.3M followers
Justin Bieber
@justinbieber
91.8M followers
KATY PERRY
@katyperry
90.1M followers
Taylor Swift
@taylorswift13
84M followers
Lady Gaga
@ladygaga
75.5M followers
Virat Kohli
@imvkohli
73.5M followers
Kim Kardashian
@kimkardashian
70.9M followers
YouTube
@youtube
68.8M followers
Neymar Jr
@neymarjr
66.5M followers
Bill Gates
@billgates
65.3M followers
Selena Gomez
@selenagomez
63.1M followers
The Ellen Show
@theellenshow
62.2M followers
CNN
@cnn
61.7M followers
X
@x
60.7M followers