Ted Spare retweeted
New Post: Designing an Always-On Agent
How to build agents like Grok Bot, Muse and Instinct
rubriclabs.com/blog/always-o…
Ted Spare retweeted
Why don’t today’s agents feel like Jarvis?
Before ChatGPT, we imagined AI would be always-on, proactive, and personalized.
But today’s AI isn’t like this. We spin it up for tasks, and it only knows the context we give it.
So what does it take to build an always on agent?
1/ Single threaded.
All messages go to one agent, no threads, no teams. This gives the agent a coherent identity. The hard part is managing the context window.
Practically, give the agent:
→ tools to fold and drop its own context so it preserves its momentum.
→ sub-agents to dispatch heavy work to, so the main thread stays free.
2/ Proactivity.
Your agent should be able to do work without being prompted and send you messages as needed. Architecturally, de-syncing messaging and awakeness, by giving the agent:
→ a heartbeat that wakes it up on a schedule.
→ tools to set its own cadence and reminders.
→ tools for sending and reading messages.
3/ Situational awareness.
Your agent should act on partial information, fill in the blanks, and kick off work from signals that didn’t come directly from your messages.
Give the agent
→ broad access to your file system, email and accounts.
→ an inbox that polls and receives webhooks, so it has real-time information.
4/ Discernment.
The hardest part of making an always-on agent good is teaching it when to act.
Today’s models are trained to crush tasks, so they react to minor notifications, get stuck doing busy work, and miss real chances to help. Ideally, your agent could start out shy, and over time build the confidence to be proactive.
This involves giving the agent:
→ a clear sense of what it’s responsible for, and that doing nothing is fine.
→ permission to ask when unsure about what to do.
→ guidelines for how to use memory and accumulate intuition
rubriclabs.com/blog/designin…
Ted Spare retweeted
⌘⌥✗⚭✲≠
Meet Common. Every symbol your keyboard is missing, built for the Mac.
No login. No subscription. Free forever.
Available at common.wtf
Ted Spare retweeted
On the Rubric blog you can actually try out these demos yourself!
We think a lot about how to share ideas as accessibly as possible without wasting the attention of our technical audience
We want our blog to be easy to skim and handle deep attention if you decide to pull over
What can you build with Jev?
Practical use cases for fast, cheap decision models.
rubriclabs.com/blog/what-can…
Ted Spare retweeted
What can you build with Jev?
Practical use cases for fast, cheap decision models.
rubriclabs.com/blog/what-can…
Ted Spare retweeted
Ollama now supports Jev-like decision models all locally.
Use decision models like Nimble for tasks like ticket triaging, model routing, and content moderation.
ollama pull nimble
Here’s Nimble playing Ollama racer through the new local /v1/systemone API by making decisions in real-time. 🏎️
Ted Spare retweeted
I've been reflecting on this. I'm generally a very optimistic person, but lately I've been feeling a strange sense of sadness that coding has been "solved".
I can, and have, made many caveats to "solved"... but I also know that in five years none of them will matter. It will be solved.
Building things is still fun, but a different kind of fun. It doesn't activate the same parts of my brain as trad coding did.
I think it's okay to mourn the thing we loved doing being different now. Software was the first job to go through the five AI stages of grief, which means we're also some of the first to internalize that this will happen to many other jobs.
It was the best of times, it was the worst of times.
Ted Spare retweeted
Just launched MapBench with @RubricLabs
We built Cartograph to generate deterministic structural representations of repositories, then benchmarked whether they help agents navigate and solve real software engineering tasks.
map-bench.xyz/
Can statically-generated maps of codebases help coding agents achieve better outcomes?
We tested three methods of mapping codebases and saw modest gains on DeepSWE.
rubriclabs.com/blog/static-a…
Ted Spare retweeted
A harness is the software that makes an AI system agentic.
Agent = Model + Harness
Modern harnesses (pi, hermes, claude code) have strong opinions about:
• compaction
• memory
• planning
• dispatch
• tool calling
• verification
• permissioning
Ted Spare retweeted
S tier quality writeup from the Rubric team!
As an agent builder, this is great and compresses pretty much everything you need to know to understand how these systems actually work.
Ted Spare retweeted
The knowledge that makes AI useful is diffused. It lives with scientists, engineers, clinicians, firms. For AI to benefit from distributed knowledge, it must itself be distributed. Agree with Jensen that this is a future worth building.
For my first post, I’m sharing a letter @NVIDIA signed on why open models matter.
AI will transform every industry, power every company, and be built by every country.
Open models strengthen safety and cybersecurity, accelerate innovation and diffusion, and enable sovereignty.
The world needs both frontier closed models and frontier open models.
images.nvidia.com/pdf/Open-W…
Ted Spare retweeted
build → critique → improve → repeat
banger blog post from @TedSpare about autonomous iteration loops for coding agents
What is polish? When we improve a feature, are we flexing intelligence or exercising sustained attention?
We tested whether a critique-improve loop closes the gap between "it works" and "it's good". Across our experiments, the answer is yes.
New post: rubriclabs.com/blog/autonomo…
Ted Spare retweeted
and it's official!
we're running an AI safety hackathon this June 11-18, open to students, researchers and devs from all backgrounds
a special thanks to @redwood_ai, @rootlyhq and @RubricLabs for your support!
luma.com/ab2xlas3
Ted Spare retweeted
New post: Primitives over Pipelines
The AI systems most teams are shipping today were designed for dumber models. Now that frontier intelligence can follow instructions, reason, and self-correct the move is to give agents primitives and let them cook.
rubriclabs.com/blog/primitiv…
