Perception @Waymo | MS CS @GeorgiaTech

Sunnyvale, CA
Joined March 2018
Mukul Khanna retweeted
Yesterday at @ComputerHistory, the @Google Orchestra was a vibe 🎻 ✨ The robot conductor was the cutest surprise; they were calling it the “semi-conductor 2.0” 😂 Meeting @AmecaTheRobot was also amazing; people spoke to her in many languages, and her expressions, eye contact, and reactions felt really human-like! – at Mountain View, CA
1
1
3
108
Mukul Khanna retweeted
A brand new day: Yosemite edition 🏞️ I’ve been loving all the spider man inspired end credits, so here’s my take for one of the most special places I’ve ever visited.
4
2
1
3
440
Mukul Khanna retweeted
‘Healthy’ discussions at N of 1: padel + personalised healthcare meetup! 🧬🎾 The future of healthcare is definitely (AI) personalization + convenience. Great time chatting with folks at @GutGutGoose @IllumeLabsInc @Ronan_RX and learning about the cool stuff they’re building
1
1
2
10
1,651
Mukul Khanna retweeted
SF, you’re kinda pretty – at San Francisco, CA
5
2
21
1,393
Mukul Khanna retweeted
Day out at @Waymo 💚💙
1
2
3
305
Proud of you bhai!
hey @FarzaTV, akshat here. would love to be part of the team. @heyclicky
2
64
Excited to present GOAT-Bench: A Benchmark for Multi-modal Lifelong Navigation with @RamRamrakhya and other collaborators at #CVPR2024! 🤖 Come visit our poster at booth 168 in Arch 4A-E! 👋🏼 mukulkhanna.github.io/goat-b…
🚀 Introducing our #CVPR2024 paper GOAT-Bench 🐐! A Benchmark for Multi-Modal Lifelong Navigation. GOAT-Bench evaluates embodied navigation agents under: 🧠 Lifelong learning: Unlike episodic benchmarks, agents are tasked with navigating to multiple goals where it should leverage past experiences to be efficient. 📷 Multi-modal open-vocabulary: Tasks agents with navigating to both seen and unseen objects specified via language, category name, or image. 1/5🧵
1
6
431
Mukul Khanna retweeted
Announcing Habitat 3.0, simulating humanoid avatars and robots collaborating! - Humanoid sim: diverse skinned avatars - Human-in-the-loop control: mouse/keyboard or VR - Tasks: social navigation and rearrangement Over 1,000 steps per second on 1 GPU for large-scale learning!
8
85
15
477
261,573
🎙️ Excited to share our latest blog post: "Why everyone needs (and wants!) a personal AI life copilot"! A revolution is happening: tens of millions of people have started using ChatGPT, Sam Altman is touring the world like a rockstar. 🕺 (1/n)
2
8
1
21
3,393
The future of robot butlers starts with mobile manipulation. We’re announcing the NeurIPS 2023 Open-Vocabulary Mobile Manipulation Challenge! - Full robot stack ✅ - Parallel sim and real evaluation ✅ - No robot required ✅👀 aihabitat.org/challenge/2023…
4
67
18
308
179,096
By the time I've actually got a draft together, I've thought about the topic so much that everything I'm saying seems so obvious as to be not worth writing down anymore.
4
2
125
13,606
Mukul Khanna retweeted
Have the courage to double down on your convictions.
144
3,018
270
56,959
4,309,162
Mukul Khanna retweeted
you can simulate all sorts of dynamical systems just using photoshop filters. noise and de-noise, repeated again and again, creates bands of spreading colours
6
75
5
842
109,809
When you team up with the right Postdoc for your PhD
144
888
203
11,086
Mukul Khanna retweeted
[Re-sliced] Humans of AI Season 2 – Week 5! Releasing videos containing answers from all guests to the same question. All thanks to the efforts of @mkulkhanna! Answers to questions 29-35 👉 humanstories.ai/resliced_s2/
2
6
Mukul Khanna retweeted
Introducing the Re-sliced version of Humans of AI (Season 2)! We are releasing videos that contain answers from all guests to the same question. All thanks to the efforts of @mkulkhanna! Answers to questions 1-7 👉 humanstories.ai/resliced_s2/
1
4
2
30
Mukul Khanna retweeted
One of my biggest growth moments as a programmer was realizing that libraries I use are just code, and I could read them directly rather than puzzling it out from the docs. Even today, I am surprised how much faster I move every time I start reading a layer I'm building on.
56
240
49
3,112
If you see me give a talk about my lab’s research, all animated and without using any notes, you might think I have a natural knack and love for public speaking. I *do*, but it wasn’t always that way. 🧵
13
75
21
608