naba-Axis-pu (✱,✱)

151 posts

naba-Axis-pu (✱,✱) banner
naba-Axis-pu (✱,✱)

naba-Axis-pu (✱,✱)

@NJakecon92504

Contributor @AxisRobotics

Katılım Kasım 2025
226 Takip Edilen46 Takipçiler
Sabitlenmiş Tweet
naba-Axis-pu (✱,✱)
naba-Axis-pu (✱,✱)@NJakecon92504·
axis is creating the infrastructure that physical ai actually needs. human interactions become signals, signals become training data, and training data improves robotics in the real world. this is how intelligent machines learn beyond simulations. @axisrobotics
naba-Axis-pu (✱,✱) tweet media
English
17
3
23
200
naba-Axis-pu (✱,✱)
naba-Axis-pu (✱,✱)@NJakecon92504·
axis is creating the infrastructure that physical ai actually needs. human interactions become signals, signals become training data, and training data improves robotics in the real world. this is how intelligent machines learn beyond simulations. @axisrobotics
naba-Axis-pu (✱,✱) tweet media
English
17
3
23
200
Sam_wolf 🐺
Sam_wolf 🐺@Sam_wolf1122·
Why is everyone talking about @axisrobotics? 1️⃣ The Tech: A 6-axis control stack for total deterministic movement. 2️⃣ The Lead: Dr. G, a veteran in industrial kinematic control. 3️⃣ The Backing: $5M seed round led by Galaxy Here is a professional breakdown 🧵👇
Sam_wolf 🐺 tweet media
English
12
0
11
115
Just Lυмму
Just Lυмму@iamlummyjay·
while touching grass, let’s learn a little about AXIS everyone talks about AI robots like the hard part is the model. Is it really? the bottleneck is data. LLMs had the internet. physical AI has almost nothing. that’s the gap @axisrobotics is trying to solve @iamlogtun
Just Lυмму tweet media
English
28
1
31
238
Rahvana.182
Rahvana.182@0xRahvanaa·
Yang ngerasa benda benda di task beda beda atau gak cocok itu benerrr Tapi, itu bukan bug ternyata wkwkkw Emang dibuat gitu Axis pake sistem In-Task Randomization biar robotnya tambah belahar dan pinter pake macem macem data. Terus apa itu In-Task Randomization? Biasanya, dalam sebuah task, objek yang muncul selalu sama persis dengan gambar referensi. Dengan fitur ini, aset visual (warna, bentuk, bahan) dan tata letak (posisi objek) akan berubah secara acak setiap kali tugas dimulai. Jadiii, gausah protes lagii yaa kalo beda itemnya wkwkwkw
Axis Robotics@axisrobotics

Noticed that object descriptions and reference images sometimes don't perfectly match the actual assets spawned in a task? This is because we’ve rolled out in-task randomization to increase data diversity and improve model generalization. The actual assets may vary, but the task goal always remains the same. Please focus on the task goal rather than the specific assets. Rich scenarios, diverse combinations of atomic skills, and extensive in-task randomization build the diversity of our data.

Indonesia
37
11
41
423
Kalito |🌋
Kalito |🌋@winterfarmerK·
“If you desire to understand intelligence, do not merely study thought, study interaction.” Had Leonardo da Vinci lived in the age of robotics, he probably would’ve obsessed over one thing above all else: Data. Not abstractly. Not theoretically. But data gathered
Kalito |🌋 tweet media
English
13
0
12
74
Vishnu
Vishnu@VishnuCrypx·
𝐖𝐡𝐲 𝐢𝐬 𝐃𝐚𝐭𝐚 𝐭𝐡𝐞 𝐁𝐚𝐜𝐤𝐛𝐨𝐧𝐞 𝐨𝐟 𝐑𝐨𝐛𝐨𝐭𝐢𝐜𝐬? --> Currently , we are talking about how Ai Can Fast Learning Everything Now a Days . Ex - : (LLM ) Large Language Model has Learning Billion web pages , Books ,videos , other Ralated Real world Interaction By Information Data Learn Via Internet. --> However, robots do not learn merely by "reading." For them to learn, they require doing. -> Physical interaction data is crucial for a robot: -> How much force to apply when lifting an object -> How to maintain balance while walking -> How to react if an object slips or breaks -> How to handle real-time environmental changes --> This constitutes the biggest difference between AI models and robots. --> Internet data teaches AI to understand language. But physical-world data teaches robots to understand the real world. --> Every movement a robot makes, every sensor reading, every camera frame and even its mistakes transform into learning data. --> In robotics, data is not merely information. It is the "experience" gained by a robot. --> In the future, the companies that will dominate the field of robotics won't be those with just better models- but rather those with superior real-world interaction data. @axisrobotics @iamlogtun
Vishnu tweet media
English
14
1
21
287
Kalito |🌋
Kalito |🌋@winterfarmerK·
Every dataset we feed into @axisrobotics is another step toward smarter machines, safer systems, and a future where technology works better for humanity. The future of robotics isn’t built by one person, it’s trained by all of us. Be part of the data shaping tomorrow.
Kalito |🌋 tweet media
English
18
2
18
123
Agoesk.somi
Agoesk.somi@0xagoes1·
Physical AI bukan sekadar kecerdasan buatan ini adalah teknologi yang bisa bergerak, bekerja, dan membantu manusia di dunia nyata. Mulai dari robot, drone, hingga mesin pintar, semuanya dirancang untuk membuat pekerjaan lebih cepat, aman, dan efisien. @axisrobotics
Agoesk.somi tweet media
Indonesia
36
8
38
315
EnpiciFi (✱,✱)
EnpiciFi (✱,✱)@masssnpc·
Skala memang penting, tapi robot tidak belajar dari data mentah begitu saja. @axisrobotics berusaha membuat trajectory yang sudah dibersihkan, dirapikan, dan dibuat cukup stabil untuk benar-benar dipakai.
EnpiciFi (✱,✱) tweet media
Axis Robotics@axisrobotics

We are now open-sourcing the AxisDataCleaning pipeline. Github repo: github.com/AxisAIOrg/Axis… Browser teleoperation is one of the most scalable paths for robot data generation. Raw human input, however, is not yet model-ready: ▪️ Idle pauses ▪️ Micro-jitters ▪️ Low & Variable frame rates Raw web data alone is not enough for reliable policy training. Here is how our backend turns noisy human demonstrations into usable trajectories for downstream policy training. 🧵👇

Indonesia
40
11
44
430
nameless (✱,✱)
nameless (✱,✱)@therealtrueog·
entah kenapa task nya akhir-akhir ini makin nyebelin ya, tapi entah knpa kalo ga di kelarin kayak ada yang kurang aja rasanya, ada yang sama ga?. btw skor kalian berapa? kok punyaku stuck disini (bertanya dengan nada santai)
nameless (✱,✱) tweet medianameless (✱,✱) tweet media
Indonesia
41
11
47
361
naba-Axis-pu (✱,✱)
naba-Axis-pu (✱,✱)@NJakecon92504·
axis isn’t building another ai app. they’re building infrastructure for physical intelligence. every human interaction becomes signal. every signal trains adaptation. every adaptation improves real-world robotics. this is how machines learn the real world. @axisrobotics
naba-Axis-pu (✱,✱) tweet media
English
52
12
56
354
Agoesk.somi
Agoesk.somi@0xagoes1·
Ketika kecerdasan digital mendapat bentuk fisik menciptakan teknologi yang lebih cerdas, adaptif, dan siap membantu kita di dunia nyata. @axisrobotics @AxisRoboticsID
Agoesk.somi tweet media
Indonesia
35
9
36
246
ADL sandi
ADL sandi@purwakarta732·
Precision. Automation. Mission accomplished. 🤖✅ Our robotic arm successfully completed Task #1: “Reposition The Toy Stove” with real-time camera monitoring and seamless data upload. The future of AI-powered robotics is here. @AxisRoboticsID @axisrobotics #axisrobotics
ADL sandi tweet media
English
37
8
39
278
Samuel
Samuel@samnvendr·
just tried Axis Robotics and this project actually feels different bukan cuma klik-klik biasa, tapi beneran kayak ngelatih AI robot buat ngerti gerakan & object placement 
task-nya juga satisfying dan makin lama makin seru buat dikerjain. why i keep doing it: •simple task •futuristic vibes •potential early user advantage 👀 •fun buat diisi pas gabut kalau AI + robotics bakal gede beberapa tahun lagi, project begini jelas menarik buat dipantau
Samuel tweet media
Indonesia
37
9
43
1.9K
naba-Axis-pu (✱,✱)
naba-Axis-pu (✱,✱)@NJakecon92504·
@axisrobotics this creates a continuous loop: humans contribute → ai learns → robots improve → real-world impact grows. more signals mean smarter and more adaptive systems. @axisrobotics
English
33
8
38
171
Axis Robotics
Axis Robotics@axisrobotics·
We are now open-sourcing the AxisDataCleaning pipeline. Github repo: github.com/AxisAIOrg/Axis… Browser teleoperation is one of the most scalable paths for robot data generation. Raw human input, however, is not yet model-ready: ▪️ Idle pauses ▪️ Micro-jitters ▪️ Low & Variable frame rates Raw web data alone is not enough for reliable policy training. Here is how our backend turns noisy human demonstrations into usable trajectories for downstream policy training. 🧵👇
Axis Robotics tweet media
Axis Robotics@axisrobotics

Over a month ago, we open-sourced our task generation pipeline. Since then, we’ve run countless trials, hardened the infra, and are getting ready to ship the full stack. Here is a raw look at our asset generation pipeline 🧵

English
122
66
238
19.6K