Ben Weis (@bweisdev) header
Ben Weis (@bweisdev) profile picture

Founder of Springbird | Ex-Rippler, Googler, Cisco Meraki | I build incredible experiences

Manhattan, NY
Joined July 2024
Photo by Ben Weis (@bweisdev)Photo by Ben Weis (@bweisdev)Photo by Ben Weis (@bweisdev)Photo by Ben Weis (@bweisdev)Photo by Ben Weis (@bweisdev)Photo by Ben Weis (@bweisdev)Photo by Ben Weis (@bweisdev)Photo by Ben Weis (@bweisdev)Photo by Ben Weis (@bweisdev)
Ben Weis retweeted
Effect 4.1 has been released! New crypto modules, HttpApi improvements, and lots of other great additions and fixes. Read more here: effect.website/blog/releases… Enjoy! @EffectTS_
7
31
3
359
13,340
Swapped tsc for effect-tsrs on Springbird's monorepo typecheck. turbo run typecheck, 210 tasks, no cache: ⏱ 88s → ~45s (−48%) 💻 CPU time −46% 🧠 Peak memory 40 GB → 24 GB (−40%) Same checks, half the wait. Huge thanks to @EffectTS_ and @yielded_dev
3
153
Foldkit’s Blocks + Effect Layers preview release is incoming. Should be ready within the next few hours. (This is a test of whether 6.1 Sol gets it over the finish line or stalls waiting for user input. The user will be asleep.) Preview status and installation instructions are in the PR description. Feedback greatly appreciated! ❤️ github.com/foldkit/foldkit/p…
5
4
2
72
3,119
Ben Weis retweeted
Did it. Effect plugin included too, so no more patch packaging :) teapawt.net/theo/status/2107789940…
Announcing tsc-rs (aka ts-rust), my complete rewrite of the Typescript compiler, type checker, and LSP, all in Rust. This is a project I've had agents working on for 5 months now. I burned ~$400k of Codex tokens and got nowhere. Burned ~$20k of Opus and got there in 2 weeks. I have not read a single line of the code. tsc-rs is an open source drop-in replacement for tsc, and it's available now.
4
3
2
72
14,269
My current favorite tech conspiracy theory: @Microsoft and @github actively despises being the unpaid backbone of the JS ecosystem. The slow enshittification of @npmjs isn't incompetence—it's a calculated psyop. They are desperately trying to annoy the community into building a viable alternative so they can finally sunset the registry, wash their hands of node_modules, and be free. @MichaelArnaldi is playing right into their hand.
1
1
6
634
Ben Weis retweeted
Replying to @EffectTS_
@EffectTS_ Effect 4.1.0 when?
1
8
53
Ben Weis retweeted
hey @1Password you should have 2 KPIs: - avg credentials per workspace - avg time to fill a password please just optimize for that it's starting to get crazy how often i have to reauthenticate, unlock, or reopen the app because it's stuck
49
18
23
1,000
314,271
Ben Weis retweeted
Dear @1Password please remove the 0.5s animation everytime I unlock 😂
hey @1Password you should have 2 KPIs: - avg credentials per workspace - avg time to fill a password please just optimize for that it's starting to get crazy how often i have to reauthenticate, unlock, or reopen the app because it's stuck
61
18
5
1,024
66,136
Claude just had a “no way that just compiled on the first go” moment
1
42
Effect v4 is here. One ecosystem. Zero dependencies. The next chapter of Effect and the foundation for building reliable software and AI agents in TypeScript.
81
321
125
2,313
383,461
Ben Weis retweeted
In case this is useful for anyone else, I created this tool to inspect my @EffectTS_ projects today. npmjs.com/package/effect-ins…
1
4
1
75
3,900
One of my favorite PRs that I’ve put up in the past few months was a one shot migration from SST to @alchemy_run Complete IaC swap in <6 hours.
Taking your advice @samgoodwin89. May the best technology win
1
1
7
726
Had an awesome time with the rest of the Springbird team! Thanks for hosting an awesome event!
The NYC @EffectTS_ community was out in force tonight! Huge shout out to @zshrm and Hyper for hosting us in their amazing office!
1
13
832
The NYC @EffectTS_ community was out in force tonight! Huge shout out to @zshrm and Hyper for hosting us in their amazing office!
Thrilled to be hosting the NYC @EffectTS_ meetup with @thefubhy @MichaelArnaldi and @maxwellbrown!
6
1
34
3,011
V8 isolates & QuickJS is *not* secure by default Seeing a lot of mistakes like: - Running in the same process as your backend → one V8 CVE reads your API keys - Exposing 𝙳𝚊𝚝𝚎.𝚗𝚘𝚠() → Spectre - Removing 𝙳𝚊𝚝𝚎.𝚗𝚘𝚠() but keeping network → the network is a clock → Spectre x2 - Guest 𝚏𝚎𝚝𝚌𝚑() → SSRF to unauthed internal services - Not whitelisting egress → botnets - Traversal & symlink escapes in filesystem API - Timeouts that can't actually kill 𝚠𝚑𝚒𝚕𝚎(𝚝𝚛𝚞𝚎){} - Bad OOM guards → try 𝚗𝚎𝚠 𝙰𝚛𝚛𝚊𝚢(𝟷𝚎𝟿) We put a lot of time into security with agentOS & Secure Exec (WebAssembly-based Linux sandbox) Instead, we follow the Cloudflare Workers & Chromium model. Some functionality we put in place: - Run isolates in a separate sidecar process - Spectre timing mitigations - Virtual operating system → full Linux that never touches the host - Event pump that can preempt a hot loop - OOM contained to VM, bounded limits, & monitoring for everything - Deny-by-default egress, explicit allowlist - (Soon) jail the sidecar process
14
8
1
155
12,855
Excessive bikeshedding is just a software engineer's filibuster
43
How long til this is the second best Postgres full-text search index, behind only TIN?
I've never written a Postgres extension. I know just enough Rust to be dangerous. Anyway, here's Stannum, a full-text search extension designed to be TINQL-compatible. Honest AI slop, with tests. github.com/TeamSpringbird/st… 1/4 🧵
1
2
25
1,741
I've never written a Postgres extension. I know just enough Rust to be dangerous. Anyway, here's Stannum, a full-text search extension designed to be TINQL-compatible. Honest AI slop, with tests. github.com/TeamSpringbird/st… 1/4 🧵
1
1
10
2,293
Then I put AI in a loop: find something slow, try a change, check correctness, benchmark again. It's much faster than Lead in local tests now. How close to TIN? I don't know yet. Still plenty to measure and plenty to break. 3/4
1
1
120
Thanks @PlanetScale, and @p_reynolds for Lead and the writeups that got this going. Stannum is open source, very alpha, and I'll keep working on it. Contributions, demanding workloads, and scrutiny of the benchmarks are welcome. 4/4
3
119