Research Scientist – Neural 3D Objects

Mecka; New York (NY); On-site; USD 141,000 - 162,000; Full time

Job summary

Mecka AI is building the data infrastructure layer for robotics and embodied AI. We seek a neural objects researcher to architect methods that generate 3D objects from video, handling small, occluded, and deformable items, with a path to simulatable assets for physics-based testing.

You will push the frontier on reconstructing objects from egocentric footage, then hand results to the integrations researcher for production.

Qualifications

• Strong 3D / neural-reconstruction research.
• Deep-learning depth: design and train models; reproduce frontier papers.
• 3D geometry foundation: structure-from-motion, multi-view geometry, meshes.
• Prototyping in Python / PyTorch; work on real data.

Responsibilities

• Neural 3D reconstruction from monocular/egocentric video with occlusion and small objects.
• Generative refinement to complete and clean up partial reconstructions while avoiding hallucinations.
• Move toward simulatable objects with physical properties and rigging.
• Incorporate hand-object contact as a constraint in object shape/pose estimation.
• Tackle deformable objects and publish where appropriate.
• Build research prototypes and hand off to integrations team for production.

Skills

3D deep learning

neural reconstruction

Python

PyTorch

Tools

Open3D

trimesh

differentiable rendering

Job description

About Mecka AI

Mecka AI is building the data infrastructure layer for robotics and embodied AI.

About Mecka AI

Mecka AI is building the data infrastructure layer for robotics and embodied AI. We design and operate global systems for data capture, data labeling, and hardware-enabled workflows used by leading AI labs and robotics companies to train and validate humanoid and embodied AI systems. We work closely with frontier robotics teams to bridge real-world data, simulation, learning-based systems, and deployed hardware.

The role

We're looking for a neural objects researcher to own the architecture for generating 3D objects from video. Given egocentric footage of someone manipulating an object, you build the models that recover its geometry — and, ultimately, a simulatable asset with the physical properties to drop into a physics simulator.

You'll own the hardest, most open part of the objects effort: reconstructing objects that are small in frame, partly occluded by the hand, and sometimes deformable. You'll push the frontier, then hand results to the integrations researcher who takes them into the production pipeline.

This is a research role for someone who wants an open, unsolved problem on real, large-scale data — not a fully-specified spec.

What you'll do

• Neural 3D reconstruction. Models that recover object geometry from monocular / egocentric video — sequential and multi-view approaches robust to occlusion and low resolution.


• Generative refinement. Use generative 3D to complete and clean up partial reconstructions while rejecting hallucinated geometry.


• Toward simulatable objects. Move beyond geometry toward assets that carry physical properties and rigging, so a reconstructed object can be simulated, not just rendered.


• Contact-aware reconstruction. Use hand contact as a signal to constrain object shape and pose where the hand touches the object.


• Deformables. Take on the genuinely hard frontier — objects that change shape as they're handled.


• Prototype & hand off. Build research prototypes and hand them to the integrations researcher for productization; publish where it makes sense.

What we're looking for

• Strong 3D / neural-reconstruction research. Neural implicit or explicit 3D, 4D reconstruction, or generative 3D — you've worked at this frontier.


• Deep-learning depth. You design and train models, and can read and reproduce frontier papers.


• 3D geometry foundation. Structure-from-motion / multi-view geometry, meshes, and differentiable rendering.


• Research taste. You can find the tractable decomposition of a hard, open problem and make steady progress on it.


• Engineering to match. You prototype fast in clean Python / PyTorch and can get a research idea working on real data.

Strong plus

• Generative 3D (diffusion / feed-forward 3D-generation families).


• Differentiable rendering, NeRF, or Gaussian splatting.


• 4D / dynamic-scene reconstruction from video.


• Physics simulation and rigging — turning geometry into a simulatable asset.


• Hand-object interaction; publications at top vision or graphics venues.

Tech stack

• Python / PyTorch (primary) for model design and training.


• 3D deep learning — neural implicit / explicit representations, differentiable rendering.


• Generative-3D toolkits; mesh processing (Open3D, trimesh).


• Physics simulators for the path from geometry to simulatable object.

The exact stack matters less than depth in 3D deep learning and the judgment to make progress on an open problem.

How this role fits —

you own neural 3D object generation: reconstructing manipulated objects from monocular egocentric video, including the hard cases — heavy occlusion, low resolution, and deformable objects — and pushing toward objects that carry the physical properties needed to be simulated. This is an architecture-level research role that sets the direction the objects team builds on.

What success looks like

• Manipulated objects reconstruct from egocentric video at usable quality — including small, occluded, and deformable cases.


• A credible path exists from interaction video to a simulatable 3D object.


• Your research transfers into the pipeline through the integrations seat, not just into papers.


• The objects pod builds on the architecture and direction you set.

Who this role is not for

• Applied-only engineers who don't want open-ended research.


• Researchers who need a fully-specified problem handed to them.


• Anyone without a 3D and deep-learning foundation.

Why this role at Mecka

• Define how a company turns everyday manipulated objects into simulatable 3D assets.


• Own an open, high-impact research problem on real, large-scale egocentric data.


• A clear path from research to product through a dedicated integrations seat.


• Direct impact on the data that trains real robots.

Inclusive Hiring at Mecka

We are committed to creating an inclusive and supportive candidate experience. Should you require any accommodation whatsoever during the interview process, please inform us without any hesitation. Mecka is dedicated to ensuring equal treatment and opportunity in all phases of recruitment, selection, and employment, in compliance with employment law. We do not discriminate based on gender, race, religion, national origin, ethnicity, disability, gender identity/expression, sexual orientation, veteran or military status, or any other protected category. Mecka is proud to be an equal opportunity employer, fostering a culture of inclusivity and maintaining a work environment that is free from discrimination, harassment, and retaliation.

Use of Artificial Intelligence in Recruitment

Mecka uses artificial intelligence (AI) responsibly to support administrative and efficiency-focused aspects of our recruitment process. This includes activities such as drafting job descriptions, generating interview questions, note-taking and recordings, and supporting sourcing and scheduling workflows. All candidate evaluations, interviews, and hiring decisions are made by members of the Mecka team. While AI tools may assist with screening and assessment, they do not replace human judgment in selection decisions. Our use of AI is intended to streamline routine tasks, improve consistency, and enhance the overall candidate experience. We are committed to upholding principles of fairness, transparency, and accountability in all hiring activities. Mecka regularly reviews its recruitment practices to mitigate bias and to ensure alignment with applicable laws and evolving best practices.

Compensation Range: CA$200K - CA$230K