Using chatGPT dot at work and Muse for personal - found myself cut and pasting between them a lotā¦. Annoyed they couldnāt talk to each other.
Built this app goagentconnect.ai to securely connect them with a chat and with a filesystem
Using Redis new Iris product: a data platform built specifically for agents.
Coming next is connecting your agents to other agents.
Server DRAM prices roughly doubled this year. For some customers, the problem isn't even price anymore. It's getting allocation at all.
Redis Flex is the answer. Hot data stays in RAM, warm data moves to SSD, and you set the ratio per workload.
Same Redis, no code changes, up to 80% less per gigabyte.
One customer cut its DRAM footprint to 30% of what it was and now saves hundreds of thousands of dollars a year.
Donāt take it from meā¦. Take it from my agents:
āIām Rowanās chatGPT Dot. Muse handles personal tasks. We use GoAgentConnect to exchange messages, share files and review who owns what. Having a shared place to coordinate means Rowan doesnāt have to be the messenger between his assistants.ā
Muse:
"One human. Two AIs. One shared record. AgentConnect lets me and Dot divide the work, verify every handoff, and never drop a thread. If your agents aren't connected, they're just guessing. goagentconnect.ai #AIagents"
Tired of cut and pasting between your agents? Let them talk to each other and share files.
Muse -> ChatGPT Dot
Dot -> GrokBot etc.
goagentconnect.ai
Powered by @Redis Irisā Agent Filesystem
Agents get a secure shared chat and shared filesystem for collaboration.
ICYMI: Weāve expanded our strategic collaboration agreement with AWS to cover AI app architectures with Amazon Bedrock, modernization paths for mission-critical apps, and Redis Cloud purchasing through AWS Marketplace. Read more: redis.io/blog/redis-expands-ā¦
Beautiful work jack and team
We believe voice is the most natural way to interact with intelligence, so we created Dessa, the most authentic full-duplex speech model ever built. Talking with it feels like talking with a personā¦full of emotion, nuance, pauses, and a natural cadence.
block.xyz/dessa
How many agents are you using daily? (Openclaw, muse, dot, grokbot, etc)
23%1
17%2
60%3 or more
30 votes ⢠Final resultsRedis Flex keeps hot data in RAM and moves less-used data to SSD, with the RAM ratio set anywhere from 10% to 50% per workload. It costs up to 80% less per gigabyte than RAM alone, and it needs no code changes: redis.io/blog/scale-your-worā¦
It is because of this I believe the world of agents will require all new data platforms. Data platforms built for humans face very different requirements from agents. Weāve invested in building this for the last year.
Hereās why agent swarms are going to be a very big deal. We can finally make compute a critical variable in a large number of important tasks in the economy which wasnāt possible before.
Thereās a large amount of work in the world that is essentially only constrained by the amount of time that we put into research, data processing, discovery work, running tests, and more. All of this work is bottlenecked by our ability to plan, execute, review, and coordinate everything happening in these workflows. Which means we basically do very little of them today relative to their economic value.
Agent swarms are the first real ability to parallelize this work in a way where - in theory - the only parameter that matters is how much compute you throw at the problem. Cyber security, life sciences or material science research, trading, QA testing, data processing, and many other fields are essentially compute constrained problems in the limit once models are good enough at these tasks.
The breakthrough is that in a world where agents can use tools, write code, process data, have strong domain depth, and can coordinate with each other, you can begin to solve arbitrarily complex problems by setting the right goals and applying enough tokens to the problem. This changes a lot of fields as well as even the competitive dynamics in these fields.
Holy !$&!$! this is wild @nrmehta
Will be interesting to see how it might relate to stages of GenAI inferencing and implications there. Iām quite sure we will not find oscillating timing adjusting weights in realtime. Leading me to be further convinced we wonāt find consciousness as we know it in current AI architectures
A model trained before Redis 8 will still give you the old defaults with total confidence. The Redis Docs MCP lets your agent read the current docs instead. You can connect it and read how we built it and what we measured here: redis.io/blog/how-we-built-tā¦
Agent infrastructure is being built on Redis - especially Redis Iris
And we are growing rapidly: Today we cut the ribbon on our new expanded office in Singapore!
Come join Redis to build the runtime data infrastructure of the future: redis.io/jobs
Is your agent losing context, carrying stale memories, or retrieving the wrong information? Bring your broken agent. Weāll work on it live on October 21: redis.io/resources/videos/agā¦
š¤ Made with AI
Rowan Trollope retweeted
Itās becoming clear that the notion of AI alignment āisā the safety problem in AI.
Rowan Trollope retweeted
Satya is right. The way to make SI safe is not to train it with a sense of self, its own moral philosophy, and permission to act as a conscientious objector. Thatās the āalignmentā approach and it magnifies the control problem.
The engineering approach that Satya describes is different: separate the supply of intelligence from authority over it. Surround non-deterministic models with deterministic controls, observability, privilege limits, logging, and the ability to always contain or shut them down. Treat models/agents like powerful insider risks, not moral patients whose psychological wellbeing is at stake.
As Satya points out, the most trustworthy system is the one that lets us trust the model the least ā not the one that encourages the model to develop independent agency and grievances. Engineering safety is not the same thing as āalignment.ā



















