Pinned Tweet
vLLM 101
> Install vLLM, serve a model
> Predict how many users fit on your GPU
> Pick the one or two flags that matter for your traffic
> Decide from a measurement when speculative decoding, FP8, or a compile level pays
> Debug a slow or broken server
aiengineeringfromscratch.com/manual-vllm-101
> Install vLLM, serve a model
> Predict how many users fit on your GPU
> Pick the one or two flags that matter for your traffic
> Decide from a measurement when speculative decoding, FP8, or a compile level pays
> Debug a slow or broken server
aiengineeringfromscratch.com/manual-vllm-101
2
27
225
7,273
If I get this, I'll teach you to train every part of these machines on aiengineeringfromscratch.com guides.
0
2
17
1,328
RT @ghumare64: Someone literally put an entire AI engineering degree on GitHub and made it 100% free 🔥
542 lessons across 20 phases, with…
542 lessons across 20 phases, with…
0
25
0
this is insane website, who created this: aiengineeringfromscratch.com
0
32
195
15,781
RT @ghumare64: vLLM 101
> Install vLLM, serve a model
> Predict how many users fit on your GPU
> Pick the one or two flags that matter for…
> Install vLLM, serve a model
> Predict how many users fit on your GPU
> Pick the one or two flags that matter for…
0
27
0
RT @ghumare64: I seriously don't understand why schools don't use this yet.
I would've mastered Reverse Engineering in college if they had…
I would've mastered Reverse Engineering in college if they had…
0
16
0
DeepSeek’s agent training system can whip up over 5,000 sandboxes every second.
The stats in their new paper are pretty wild.
A single production unit covers about 160 nodes and manages around 3 million sandboxes a day.
At peak scale:
-> 380,000+ concurrent sandboxes
--> 5,000+-> containers, microVMs, and full VMs all behind one interface
arxiv.org/abs/2609.22978
The stats in their new paper are pretty wild.
A single production unit covers about 160 nodes and manages around 3 million sandboxes a day.
At peak scale:
-> 380,000+ concurrent sandboxes
--> 5,000+-> containers, microVMs, and full VMs all behind one interface
arxiv.org/abs/2609.22978
0
2
10
1,648
Someone literally put an entire AI engineering degree on GitHub and made it 100% free 🔥
542 lessons across 20 phases, with ~400 hours of material.
The whole curriculum is built around learning AI from first principles.
You start with the math, then implement the concepts on, tokenizers, attention, agent loops and more. Production frameworks come later, once you've already built the core mechanics by hand.
Every lesson runs these 5 steps:
Read the problem → derive the math → write the code → run the test → keep the artifact.
By the end of each lesson you have reusable: prompts, SKILLmd files, agent definitions, MCP servers you built yourself. The code runs in Python, TypeScript, Rust or Julia, depending on what fits the concept best.
The curriculum goes from linear algebra all the way to autonomous agent swarms, multi-agent coordination, production infrastructure, and AI safety.
It’s 100% free, open source, MIT licensed and available to clone and run locally. 73k+ stars on GitHub.
aiengineeringfromscratch.com
542 lessons across 20 phases, with ~400 hours of material.
The whole curriculum is built around learning AI from first principles.
You start with the math, then implement the concepts on, tokenizers, attention, agent loops and more. Production frameworks come later, once you've already built the core mechanics by hand.
Every lesson runs these 5 steps:
Read the problem → derive the math → write the code → run the test → keep the artifact.
By the end of each lesson you have reusable: prompts, SKILLmd files, agent definitions, MCP servers you built yourself. The code runs in Python, TypeScript, Rust or Julia, depending on what fits the concept best.
The curriculum goes from linear algebra all the way to autonomous agent swarms, multi-agent coordination, production infrastructure, and AI safety.
It’s 100% free, open source, MIT licensed and available to clone and run locally. 73k+ stars on GitHub.
aiengineeringfromscratch.com
0
25
229
11,463
I seriously don't understand why schools don't use this yet.
I would've mastered Reverse Engineering in college if they had thought this way.
I would've mastered Reverse Engineering in college if they had thought this way.
0
16
143
5,563
RT @ghumare64: It's very easy, learn here: aiengineeringfromscratch.com
0
92
0
RT @ghumare64: I seriously don't understand why schools don't use this yet.
I would've mastered AI Engineering in college if they had thou…
I would've mastered AI Engineering in college if they had thou…
0
88
0
1
Once you start using Grok Bot there is no going back to dots or hermes or openclaw!! Amazing job.
1
0
16
2,826
I seriously don't understand why schools don't use this yet.
I would've mastered AI Engineering in college if they had thought this way.
I would've mastered AI Engineering in college if they had thought this way.
2
88
1,135
39,119
It's very easy, learn here: aiengineeringfromscratch.com
"Building agents is not that hard."
1
92
1,154
64,433
Congrats on the launch boys! 💯
Shipping code shouldn’t mean hours of manual testing.
@Simulithic (YC F26) sends AI agents through your app like real users, across browsers, devices, and account types. It tells you what broke without you writing a single test script.
The teams that ship fastest will be the fast as they build.
Get started today: simulithic.com
0
0
11
3,069
this is taking a very good shape !!
below is the animation for @Google A2A protocol.
there are more:
aiengineeringfromscratch.com/manuals
below is the animation for @Google A2A protocol.
there are more:
aiengineeringfromscratch.com/manuals
0
2
18
2,929
RT @ghumare64: Who's gonna tell ai engineers about...
they are wasting paying for paid courses...
while they can learn:
Maths
Languages
F…
they are wasting paying for paid courses...
while they can learn:
Maths
Languages
F…
0
11
0