Eran Sandler (@erans) header
Eran Sandler (@erans) profile picture

Builder, operator and investor. Infra, AI, and product nerd. Opinions are my own. Checkout @lunaroute_ai (https://teapawt.org/t.co/YlZrgrIUKq) to stop caring about token limits

Joined April 2007
Eran Sandler retweeted
Replying to @lunaroute_ai
@lunaroute_ai added inkling-2-small earlier today, we're currently running our design harness across briefs, variants and skills and will soon share the results. While it's running we got a glimpse of the early results and boy this one seems promising for its size.
1
2
1
3
198
Eran Sandler retweeted
Introducing opencomputer.dev An open source, faster Devin alternative you can self-host 0:23 Why it's 3.3x faster 1:27 One VM is all you need 2:01 Fully self hostable 2:16 Faster + cheaper at scale 3:27 Durable + secure 3:51 Star us on GitHub!
51
45
16
328
23,707
Eran Sandler retweeted
5/ Take it for a spin (it's still being refined so share feedback!) joeysim.github.io/joeysim-sk…
1
1
10
1,103
Eran Sandler retweeted
If you're in need of dedicated inference that will cost you a fraction of what you're paying OAI or A/ talk to @erans and @lunaroute_ai . You'd be amazed how big of unlock it'll be for your teams
We will see more and more companies doing this. Not paying by the tokens enables Continuous AI for doing all sorts of things. If you can't afford 53 B200 nodes talk to us at @lunaroute_ai - we can get you private fixed cost inference and more than just GLM 5.3 for much less!
1
1
2
307
Eran Sandler retweeted
I blame @erans for creating @lunaroute_ai
1
1
3
99
Eran Sandler retweeted
1/ I've been having lots of fun lately, letting my curiosity lead me down rabbit holes. This is one experiment I've been working on - trying to answer if AI models have distinctive design taste.
1
1
5
284
Eran Sandler retweeted
5/ Credit where credit is due - @erans' project @lunaroute_ai flipped the bit for me on using AI without worrying about token cost thanks to their concurrency based billing.
1
2
3
83
Eran Sandler retweeted
Write On 1.1.0 is out: my open-source Mac take on @jasonfried's Write On app concept. New: AI trims ghost your words instead of deleting them, so any cut can be revived. Plus a model picker and a round of fixes that make the AI features more reliable. github.com/joeysim/write-on-…
1
4
213
Eran Sandler retweeted
Been using @lunaroute_ai myself and it’s changed how I use AI for coding. With unlimited tokens, I’m running way more evals and benchmarking different models. I’m much more willing to experiment when I’m not thinking about the cost of every run.
1
5
80
Eran Sandler retweeted
LunaRoute opens a mental barrier that we put on ourselves! @joeysim is a very happy early users. Join him - Free 7 days trial for Meta Muse Spark Open, GLM 5.3, GLM 5.3 Flash and DeepSeek 4.1 Flash. You owe it to your creativity!
Replying to @joeysim
2/ I was trying to debug why this happened, I have a Pro account on both Anthropic and OpenAI, so what gives. It boils down to this
2
8
1,226
Eran Sandler retweeted
2/ I was trying to debug why this happened, I have a Pro account on both Anthropic and OpenAI, so what gives. It boils down to this
1
1
1
2
1,335
Eran Sandler retweeted
We did it! Framework Laptop 13 Pro is on TIME's Best Inventions of 2026. This is our third product to list, after the original Framework Laptop in 2021 and Framework Laptop 16 in 2023.
51
90
18
2,060
32,557
Eran Sandler retweeted
we have fully switched to this new way of building internally now all work - not just product development - now gets done in Slack threads it feels FAST, and it feels RIGHT. speed matters! and ofc we are building Computer with Computer DM for access
4
3
8
586
Eran Sandler retweeted
seems obvious to everyone around me that sandboxes are on their way out. we hear too many stories about reliability, capacity, broken sandboxes that need manual intervention -- plus they are way too expensive for what they provide. the future for agents? workspaces. your agent doesn't need a computer. it needs a place to store it's context, and the *tools* to manipulate that context each turn. we're in the very early stages of the AI era. the simplest version of this pattern is just handing each agent a time-multiplexed VM, since that's the only thing that we have at hand right now. over the next 12 months, you're going to see a rapid shift away from this model and towards what i call the "magic compute box". rather than giving an agent a VM, you give it a set of tools (could be "bash", "python", "js", "read_file", "write_file") -- and you send those commands to a magic service in the cloud that manages to run them for you -- using the cheapest technology available (just-bash, wasm, or even Linux if really required) for each command invcation. your agent doesn't want a computer that it can set a desktop background for, it wants a place where it can fire off an unlimited number of tool calls and have it update the context correctly.
is this a hot take? if you're using persistent VMs with agents you're doing it wrong ephemeral sandboxes are not only the future, but they are already the present, and they provide numerous and superior capabilities to single or small quantity persistent VMs
77
15
25
390
50,283
Eran Sandler retweeted
The headline news is we have filled the gas tank again at USV and this tank is bigger. But scroll down to see the four areas we want to invest these funds into.
50
22
5
585
61,591
Eran Sandler retweeted
.claw is now a thing. Always wanted to make a TLD happen on the list of life goals. 🦞
.Claw cometh! The winning applicant for .claw top level domain in ICANN’s 2026 round: the OpenClaw Foundation. Huge step in giving our agents a natural home presence. Congratulations to @davemorin @steipete and the @openclaw team. Looking forward to great things crawling 🦞 across the web soon.
9
4
79
14,864
Eran Sandler retweeted
The @openclaw foundation has officially won the @ICANN application for .claw giving open agents a durable home at the root of the internet. 🦞
16
28
4
269
22,117
Eran Sandler retweeted
Introducing Beam: a highly efficient agentic open model with 501B total parameters and 23B active. - Frontier reasoning efficiency - Advances the Western open frontier on coding & agentic tasks - Trained end-to-end from scratch Full weights release this month. Learn more about Beam: reflection.ai/beam
564
1,087
759
8,591
2,554,692
Eran Sandler retweeted
millions concurrent is cool but how bout 200k/s agent session creation rate?
was curious about how fast we can burst create @opencomputerHQ serverless agents so I did this experiment. Hitting 4000/s sessions created at peak Each session is a full opencode session starting successfully. Create your agent swarm on Opencomputer in a few lines of code .. leave the scaling and sandboxes to us
1
2
269