bertdb

Joined 15 years ago

Karma
310
Submissions
21
Member since
Jan 13
bert AT segments.ai
View on Hacker News

Recent submissions

Comment · 5 months ago
Uber | https://www.uber.com/ai-solutions | Senior Product Engineer | Full Time | Amsterdam, The Netherlands | Hybrid

We’re building a complex data labeling platform for AV and robotics teams working with multi-sensor data: images, point clouds, sequential data, and 3D scenes.

We’re hiring a Senior Product Engineer to help us ship the core web product faster and improve the quality of customer-facing labeling workflows.

Tech stack: Vue/React, TypeScript, Three.js/WebGL/WebGPU, TinyBase, Django.

You can help us solve problems like: - Making 2D and 3D annotation workflows faster, more precise, and easier to use - Designing point cloud tiling, autosave, undo/redo, review, filtering, and issue-tracking workflows that hold up under real labeling operations - Improving performance where milliseconds matter: rendering, local state updates, preloading, and geometric operations - Shaping API contracts, data models, permissions, background jobs, and operational tooling as the platform scales

This is a senior role with high ownership on a small team. To apply, please reach out to [email protected] (we recently got acquired by Uber) with a resume and mention HN in your application.

Comment · 2 years ago
Did you do any inpainting experiments? I can imagine a pixel-space diffusion model to be better at it than one with a latent auto-encoder.
Comment · 4 years ago
Could the NeRF be replaced with a voxel grid, backpropagating to the voxel color values directly? Or is there a reason that wouldn't work?
Comment · 5 years ago
Any open source projects with a nice Django/DRF (+ Vue/React frontend) codebase, that roughly follow this style guide?
Comment · 5 years ago
Thanks! Would be great to learn more about your use case, we'll be in touch with you.
Comment · 5 years ago
Comment · 5 years ago
The segments you see on the screen are generated by our ML model. If your data is very noisy, our out-of-the-box model might not be the best fit. We can always improve performance by training a custom model for you on a small set of manually labeled data though.
Comment · 5 years ago
Thanks for your feedback!

1. If the segment you start dragging from is already selected, all the segments you drag through will get deselected, and vice versa.

2. Did you try changing the granularity of the segments by scrolling your mouse wheel? We've had good experiences with microscopic imagery before, happy to connect and dig a bit deeper.

Comment · 5 years ago
Happy to listen to what you need, feel free to shoot me an email.
Comment · 5 years ago
Thanks Brian!
Comment · 5 years ago
Looking forward to hearing your feedback when you give it a try!

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton