Random Libertarian Tech Lead (@someRandomDev5) profile picture

Shipping to prod in a $100M+ ARR unicorn startup w/ local AI. 4x DGX Spark (GLM 5.3 Flash), 1x M5 Max 128GB (Qwen 3.8 Flash Next), 1x RTX 5090 (Qwen 3.8 27B).

Joined February 2018
Photo by Random Libertarian Tech Lead (@someRandomDev5)Photo by Random Libertarian Tech Lead (@someRandomDev5)Photo by Random Libertarian Tech Lead (@someRandomDev5)Photo by Random Libertarian Tech Lead (@someRandomDev5)Photo by Random Libertarian Tech Lead (@someRandomDev5)Photo by Random Libertarian Tech Lead (@someRandomDev5)Photo by Random Libertarian Tech Lead (@someRandomDev5)Photo by Random Libertarian Tech Lead (@someRandomDev5)Photo by Random Libertarian Tech Lead (@someRandomDev5)Photo by Random Libertarian Tech Lead (@someRandomDev5)
Random Libertarian Tech Lead retweeted
“I apologize for such a long PR - I didn't have time to write a short one.” - Mark TwAIn
2
5
40
1,407
Random Libertarian Tech Lead retweeted
When Dario asks Sam to pace the frontier right before Anthropic's IPO.
🤖 Made with AI
25
41
3
620
46,724
Random Libertarian Tech Lead retweeted
8
126
9
1,983
34,000
Random Libertarian Tech Lead retweeted
The drunk "LLMs are not conscious" posting will continue until morale improves.
2
4
2
45
1,189
Random Libertarian Tech Lead retweeted
Replying to @luckeyfaraday
Anyone who vibe-benches with pagodas, 3d models, etc., proves by doing so that they have no idea whatsoever how LLMs or evals work. Anyone who does this, you should mute them so that the intelligence-per-post ratio on your "For You" feed goes up.
1
139
Random Libertarian Tech Lead retweeted
47
115
47
822
286,933
Random Libertarian Tech Lead retweeted
🚨BREAKING: First gameplay trailer for Holdout Juror Simulator. 11 against 1. Seven days. Can you hold your ground?
🤖 Made with AI
705
6,777
423
72,089
3,441,840
Random Libertarian Tech Lead retweeted
the fly brain can play beat saber
1,051
8,010
2,431
85,237
23,138,853
Random Libertarian Tech Lead retweeted
Okay, this is the best way to explain why AI Slop is not what you think. By using AI Denzel Washington to explain it all. This is actually really good. Well written, directed and generated. It's about you not needing the studios anymore.
483
1,110
215
8,500
471,925
Random Libertarian Tech Lead retweeted
Replying to @sama
> agi is here > A BLOG POST doesn't work smfh 2026 is not real
3
71
5,739
Random Libertarian Tech Lead retweeted
Local AI addiction starts innocently. I just want privacy. Then you’re downloading 200GB checkpoints at 2am, comparing 4-bit quants, calculating memory bandwidth over breakfast, and calling a new GPU infrastructure. Anyway, one last benchmark and I’m going to bed ☺️.
40
30
7
449
16,108
Random Libertarian Tech Lead retweeted
Replying to @claudeai
Finally, we’ve improved our safeguards. Our cybersecurity safeguards now flag benign requests about 60% less often. On basic biology and medical questions, we’ve recently reduced the fallback rate by around 85%.
80
239
32
6,843
381,065
Random Libertarian Tech Lead retweeted
Replying to @cremieuxrecueil
interesting that reasoning appears to pull them towards center or right
42
114
17
5,591
128,435
Random Libertarian Tech Lead retweeted
Lead Engineer, Scott Fryxell, used Pi to move most of his client work to DeepSeek, dipping into Fable only when necessary. By primarily using ‘commodity models’ he only needs two $20 plans for his total usage. Read how Pi became “the most important piece” of Scott’s rig below
41
157
22
2,516
198,308
Random Libertarian Tech Lead retweeted
2 years ago, we achieved the first big milestone with @exolabs. We clustered 2 MacBooks to run Llama 405B. It felt like magic. The consensus was running this model was only possible in a data center. We ran it on consumer hardware, on 2 M3 Max MacBook Pros. Most people thought it was a gimmick. It only ran at 2 tok/sec! But, we believed that improvements to the software, hardware, and models would all compound. So that maybe in a few years, we thought, this would improve 10x in software, 10x in hardware, 10x models = 1000x. That was the vision. We imagined a world where you would have frontier intelligence running quietly on your desk. Today is the day that vision became reality. The M5 Ultra is a 10x step-change improvement vs the M3 Max we originally clustered. That, compounded with software improvements like RDMA over Thunderbolt, MTP and better kernels, and high intelligence density models like Qwen 3.8 27B, means we now have 1,000x better Local AI than when we started. I am so grateful to the small group of people at Apple (including @doogie69 @awnihannun @angeloskath @DiganiJagrit @doogie69) who believed in this vision and had the foresight as well as the courage to take a swing at this early on. I'm confident they're just getting warmed up (looking forward to 4-bit / 8-bit compute units in M7 Ultra🤞). With the M5 Ultra Mac Studio, we are going to have unmetered tokens running at API speeds at effectively zero marginal cost, running on your desk, so quietly and consuming so little power you won't even notice it. Local AI is good now.
exo featured on Apple's new M5 Ultra Mac Studio and M6 / M5 Pro Mac Mini pages. Over the past year, we have worked closely with Apple on low-latency RDMA networking over Thunderbolt 5, enabling clusters of Macs to run massive models like Kimi K3 and GLM-5.3 at API speeds. With RDMA, aggregate memory bandwidth across Macs scales ~linearly. A cluster of 4 x M5 Ultra Mac Studios scale to an aggregate memory bandwidth of ~4.8TB/s. Previously, these were speeds only achievable with data center GPUs. Our vision is a data center on every desk. Apple Silicon's superior memory unit economics, power efficiency, and out of the box experience for Local AI make that possible. Thank you to @angeloskath, @awnihannun, @twid and countless others at Apple who tirelessly to bring this technology to the world.
49
62
12
806
77,046
Random Libertarian Tech Lead retweeted
Specialist AI’s (single language, single area of knowledge) are another 100X
Elon Musk’s January 6 Moonshots podcast prediction of 100x gains in intelligence at a fixed model size was at the edge of plausibility when made, now it’s simply a fact. What next?
731
846
118
7,910
3,838,299
Random Libertarian Tech Lead retweeted
Replying to @rezoundous
Around the time of Opus 4.6, there was a pretty clear distinction between Anthropic and OpenAI models: Anthropic: Very intelligent model, good at understanding people, their desires, motives, and intent. OpenAI: Very intelligent model, good at understanding systems.
1
3
1
1
2,284
Random Libertarian Tech Lead retweeted
Replying to @mattshumer_
1
20
289
7,313
Random Libertarian Tech Lead retweeted
All discussions around AGI are truly just semantic debates arguing about what AGI means. Every single one of them.
1
1
5
238