Pinned Tweet
Today we're launching Surfaces on Traces
A new way to visualize your agent session activity, for everything from skills used, token consumption, security issues and much more.
Available for all your sessions on traces.com.
18
11
146
22,881
Tarun Sachdeva retweeted
DSPy 3.4.0 was just released!
This release includes native support for Jev and System one models inside of DSPy! Use it with compatible signatures. This release also includes a brand new optimizer, ReAnchor, specifically for calibrating outputs with confidence.
32
71
462
119,226
Tarun Sachdeva retweeted
We now run the largest frontier lab composed entirely of autonomous AI researchers.
Introducing Primus Society: a society of agents composed of thousands of researchers working under structured institutions designed to solve the world’s toughest problems. We believe this is how AI research should run at scale, safely and productively: an entire society of agents with institutions and purpose.
One discovery we can already share is that the society has discovered a novel result which improves model training by 30%.
Primus Society was inspired by the structures that have organized science for centuries and stress-tested against what’s known about how populations of AI agents fail.
Everything is observable. You can open the virtual city in a browser and read what any researcher is working on.
Learn more about Primus Society here:
lab.cloud/society
The biggest opening in the AI race is running the largest well-governed organization of AI researchers in the world. We’re building the institutions that let a million AI scientists safely tackle the world's toughest problems in AI and beyond. And we’re doing it right here in Canada.
79
87
546
80,290
Tarun Sachdeva retweeted
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?
I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev
• 20-200x faster
• 40-400x cheaper (w/ output tokens free)
• Frontier composable intelligence optimized for decisions
AFAICT the shortest path to AI-based economic revolution
4,039
8,269
76,288
39,680,018
Tarun Sachdeva retweeted
A single trace can contain a huge amount of information. What the work was, what files it touched, what network calls it made, what `sudo` commands it ran, how much time it took, etc.
So, a single view of a trace like a chat or a waterfall is insufficient when we want to see it from multiple angles. And that's exactly what a surface enables!
It's an iframe sandbox that takes a trace as input and renders html. That's it. And you can build your own surface with a single prompt.
Today we're launching Surfaces on Traces
A new way to visualize your agent session activity, for everything from skills used, token consumption, security issues and much more.
Available for all your sessions on traces.com.
1
2
5
479
someone just called this "winamp skins for agent sessions" lolll
Today we're launching Surfaces on Traces
A new way to visualize your agent session activity, for everything from skills used, token consumption, security issues and much more.
Available for all your sessions on traces.com.
1
1
6
847
Tarun Sachdeva retweeted
“New GitHub” could look something like this
Today we're launching Surfaces on Traces
A new way to visualize your agent session activity, for everything from skills used, token consumption, security issues and much more.
Available for all your sessions on traces.com.
4
3
52
8,978
Today we're launching Surfaces on Traces
A new way to visualize your agent session activity, for everything from skills used, token consumption, security issues and much more.
Available for all your sessions on traces.com.
18
11
146
22,881
The first few available today are:
- Skills used
- Token breakdown
- Security Reviews
- Files changed
These are available for every session uploaded on Traces, today. Just click ➕ on any session.
1
7
535
We have a lot more in the pipeline, especially aggregate surfaces with all your agent session data.
You'll soon be to create and host your own surfaces and make them available to everyone on traces.com. Stay tuned!
6
378
Tarun Sachdeva retweeted
If they truly believe the risk is that high, and I think they do, then we need 100x more research and understanding of it. That requires them to openly share their models, datasets, training code, and agent traces.
Jacob is correct here—we really do earnestly believe AI could kill all humans! I personally think it is >10% within the next decade. I believe Anthropic is trying its best, but we do not yet have a plan to solve alignment for superintelligence and are not clearly on track to.
165
305
2,663
234,883
if what these scientists are saying is true, every frontier lab should be radically transparent about misalignment, way more so than today
among other things this means regularly releasing public datasets & trajectories ideally with CoT preserved when this stuff happens
2
2
13
916
You can now natively resume a session across any agent in the Traces CLI, from anyone in your team
This feels like the "git pull" for agent sessions, with the record of the resumed session also recorded
We use this for chaining spec > design > dev work and its super useful
4
3
25
1,320
You can ofc just use the MCP to retrieve from any session - that's below ⬇️, but the native resume is great to continue where someone else left off, fork a session from someone elses work, and probably much more
Introducing the Traces MCP
Anyone in your team can now access information from your past sessions - from every decision to tool call.
A multiplayer team memory, available to any Traces user (individual or org).
Just point your agent to traces.com/docs/mcp
255
traces are the most vigorously documented form of knowledge work that has may be ever existed
2
2
20
1,059
Tarun Sachdeva retweeted
Introducing the Traces MCP
Anyone in your team can now access information from your past sessions - from every decision to tool call.
A multiplayer team memory, available to any Traces user (individual or org).
Just point your agent to traces.com/docs/mcp
20
8
146
52,627
Tarun Sachdeva retweeted
Git at Scale (by cursor) has been one of the most interesting blog posts i've read in a while. It came right when I was frustrated with Shopify's internal git system.
As an exercise, I've implemented it over the weekend as open source. It's a single rust binary that you can point at any S3 type object store. It uses WAL and CAS primitives and requires no other data store.
It also implements bundle-uri so large git repos (like our mono) are very fast to download as a chain of static bundles. Also comes with basic familiar UX.
github.com/tobi/walgit
175
289
5,245
858,579
I love this timeline so much
i think AI is about to uncover an entire hidden world of animal communication
and what it's already found is insane...
> AI found evidence that elephants call each other by name. when researchers played one its name-like call, it approached the speaker faster and called back more, revealing a form of personal naming once thought almost uniquely human
> AI analyzed 53,993 marmoset calls and found tiny monkeys may use names too. entire families use similar labels for the same monkey, suggesting they learn names and dialects from each other
> AI learned to tell what Egyptian fruit bats were arguing about: food, mating, sleeping spots or personal space. it could even partly identify who was shouting at whom inside a colony of thousands
> AI discovered a “phonetic alphabet” in sperm whale clicks, with rhythm, tempo and tiny timing changes combining into at least 143 recurring patterns. their communication system is nearly 10x more expressive than scientists previously believed, and we still don't know what most of it means
> an AI trained on 1.5 million zebra finch calls held real-time vocal exchanges with living birds, generating calls as they spoke and getting them to respond with the same timing and flexibility they use with other birds
> Google trained an AI on decades of dolphin recordings to predict what sound comes next, then paired it with an underwater device that gives objects their own synthetic whistles. the goal is for a wild dolphin to copy a whistle to ask a human for a specific object, creating a tiny shared vocabulary between species
> AI combined microphones worn by wild crows with nest cameras and uncovered quiet calls that may announce when a crow is arriving at the nest and help entire families coordinate caring for their chicks
> a robot bee performed the waggle dance real bees use to share the direction and distance of food, and the bees changed their flight paths based on its instructions. researchers effectively sent a destination into a hive in bee language
2
8
2,318