GPT 5.5 powered Hermes as a PM for each project/repo, accessed via telegram….
….that prompts a Claude Code session in tmux to build out UI…
….And a codex session in tmux to do the coding
So a day driving mostly with codex 5.5 and I have to say I’m actually really impressed, but there’s a caveat—for the first time it seems genuinely to perform better within the app than the terminal. Auto-review works great (seems better again in app than parallel permission in terminal).
My gut right now: planning is better in claude (much more interactive - 5.5. seems to just think YOLO and jump in and start building). design is better in claude. coding is better in codex. fortunately this creates quite a neat divide of tasks. do planning and front-end in claude then get codex to pick up the plan and build them.
Introducing GPT-5.5
A new class of intelligence for real work and powering agents, built to understand complex goals, use tools, check its work, and carry more tasks through to completion. It marks a new way of getting computer work done.
Now available in ChatGPT and Codex.
browser-harness has changed everything for me. I now default to asking claude code, "can you drive?" for anything I don't want to do. Use Claude Design? Great idea. Can you drive?
🚨 GENERATE GPT-IMAGE-2 UNLIMITED. FREE.
uses your $20/mo ChatGPT Plus sub. no API key, no metered billing.
every coding agent — Claude Code, Codex, Cursor, OpenCode, KIMI.
brew install paperfoot/tap/nanaban
https://t.co/1E09PAle3I
Once again, say it with me
As an industry, we need to move on from static secrets
I’ve been going insane saying this shit for years, there are literally replacements in production today
re: Vercel hack
This should be our wakeup call to get rid of API keys, just like we "got rid" of passwords with OAuth & passkeys
There's a fix. Of course there's a fix! It's called OIDC Federation, it works beautifully, but it still hasn't replaced API keys in the mainstream.
In short, your infra (Vercel, AWS, etc.) generates short-lived JWTs. You then tell your services (database, AI provider, etc.) to trust those JWTs instead of the API keys. No static secrets required.
These JWTs have a lifetime of <1h, and rotate automatically. Simpler + safer
Is it time?
Love this so much. Between agent chat rooms as pixel games and this kind of thing, I definitely feel that one of the greatest things about the vibe code age is that it’s become cheap to code whimsical things
Introducing Pods
Hyperspace Pods lets a small group of people - a family, a startup, a few friends, to pool their laptops and desktops into one AI cluster. Everyone installs the CLI, someone creates a pod, shares an invite link, and the machines form a mesh. Models like Qwen 3.5 32B or GLM-5 Turbo that need more memory than any single laptop has get automatically sharded across the group's devices - layers split proportionally, inference pipelined through the ring. From the outside it looks like one OpenAI-compatible API endpoint with a pk_* key that drops straight into your AI tools and products. No configuration beyond pasting the key and changing the base URL.
A team of five paying for cloud AI burns $500–2,000 a month on API calls. The same team's existing machines can serve Qwen 3.5 (competitive on SWE-bench) and GLM-5 Turbo (#1 on BrowseComp for tool-calling and web research) for free - the hardware is already on their desks. When a query genuinely needs a frontier model nobody has locally, the pod falls back to cloud at wholesale rates from a shared treasury. But for the daily work - code reviews, refactors, research, drafting - local models handle it and nobody gets billed. And when it is idle, you can rent out your pod on the compute marketplace, with fine-grained permissions for access management.
There's no central server involved in inference. Prompts go from your machine to your pod members' machines and back: all of this enabled by the fully peer-to-peer Hyperspace network. Pod state - who's a member, which API keys are valid, how much treasury is left - is replicated across members with consensus, so the whole thing works on a local network. Members behind home routers don't need port forwarding either. The practical setup for most pods is three models covering different jobs: Qwen 3.5 32B for code and reasoning, GLM-5 Turbo for browsing and research, Gemma 4 for fast lightweight tasks. All running on hardware you already own.
Pods ship today in Hyperspace v5.19. Model sharding, API keys, treasury, and Raft coordinator are all live.
What Makes This Different - No middleman. Your prompts travel from your IDE to your pod members' hardware and back. There is no server in between reading your data.
- No vendor lock-in. Pod membership, API keys, and treasury are replicated across your own machines using Raft consensus. If the internet goes down, your local network keeps working. There is no database in someone else's cloud that your pod depends on.
- Automatic sharding. You don't configure layer ranges or calculate VRAM budgets. Tell the pod which model you want. It figures out how to split it across whatever hardware is online.
- Real NAT traversal. Your friend behind a home router with a dynamic IP? Works. No VPN, no Tailscale, no port forwarding. The nodes handle it.
- Free when local. This is the part that matters most. Cloud AI bills scale with usage. Pod inference on local hardware scales with nothing. The marginal cost of your 10,000th prompt is the electricity your laptop was already using.
Coming soon:
- Pod federation: pods form alliances with other pods.
- Marketplace: pods with spare capacity can sell inference to other pods.
Ran a test to see whether I could hook a sandboxed Hermes Agent into my various bank accounts with a python backend and a supabase db to help me track spending.
The API story in this space for the UK isn't good. @Plaid are the real leaders but "don't support personal plans" which is a missed opportunity in the agentic AI age. @TrueLayer and @Yapily are enterprise only.
@StarlingBank and @monzo have APIs, which is awesome. Unfortunately not all my banking is with them. The old banks and AMEX predictably have nothing.
Feels like there's a gap here for someone to build a b2c solution for people to get API access to their own finances.
@AlexFinn@garrytan I keep thinking about how we now all have access to the same code, and it’s our ability to use the right words that determines whether we get amazing outcomes or terrible ones. We’ve become managers, coaches, motivational experts.