# AI Research Engineer

**Company:** [Origin](https://hotfix.jobs/companies/origin-tech)
**Location:** Bengaluru, India
**Role:** AI Research
**Experience:** 3+ years
**Skills:** Python, PyTorch, JAX, Diffusion Models, Vision-Language Models, Multimodal Transformers, Active Learning, Pseudo-Labeling, Weak Supervision, TensorRT, Onnx Runtime, CUDA, C++, Ros 2, Isaac Sim
**Posted:** 2026-07-22

> Conducts research on diffusion, vision-language, and vision-language-action models for autonomous construction robots. The role requires deep-learning R&D experience or advanced graduate training, strong multimodal AI expertise, scalable data workflows, and model optimization for edge deployment.

## Job Description

## Responsibilities
- Research and innovate diffusion-based generative models for photorealistic wall-surface simulation, defect synthesis, and domain adaptation.
- Architect and train Vision-Language Models (VLMs) and Vision-Language Action Models (VLA) that connect textual work orders, CAD plans, and sensor data to pixel-level understanding.
- Lead auto-annotation pipelines using active learning, self-training, and synthetic data to scale to millions of frames and point clouds with minimal human effort.
- Optimize and compress models using INT8, LoRA, and distillation for deployment on Jetson-class edge devices under ROS 2.
- Own the full lifecycle from problem definition, literature review, and prototyping through offline/online evaluation and production hand-off to perception and controls teams.
- Publish internal technical reports and external conference papers.
- Mentor interns and junior engineers.

## Requirements
- 3+ years of deep-learning R&D experience, or a PhD/Master's degree in computer science, electrical engineering, robotics, or a related field with a strong publication record.
- Demonstrated expertise in diffusion models, including DDPM, LDM, and ControlNet.
- Experience with multimodal transformers and VLMs, including CLIP, BLIP-2, LLaVA, or Flamingo.
- Proven success building large-scale data-centric AI workflows using active learning, pseudo-labeling, and weak supervision.
- Advanced proficiency in Python, PyTorch or JAX, experiment tracking, and scalable training with tools such as PyTorch Lightning, DeepSpeed, or Ray.
- Familiarity with edge-AI runtimes such as TensorRT and ONNX Runtime.
- Experience with CUDA and C++ performance tuning.
- Strong mathematical foundation in probability, information theory, and optimization.
- Ability to translate theory into production code.

## Nice to Have
- Experience with synthetic data generation in Isaac Sim.
- Experience with robotics perception stacks, including ROS 2, Nav2, MoveIt 2, or Open3D.

## Similar jobs

- [AI Researcher](https://hotfix.jobs/jobs/077574a5-3c6d-4634-a23c-4a909ce8aa65) - Improbable - Remote
- [Research Scientist - Member of Technical Staff](https://hotfix.jobs/jobs/9807d0a2-a252-48f0-81fa-2175608c94b8) - AI Digest - Remote - $150k – $350k/yr
- [Engineer - Member of Technical Staff](https://hotfix.jobs/jobs/cba82dd2-84d4-48a7-8d75-58783156ab95) - AI Digest - Remote - $150k – $350k/yr
- [Senior Product Builder, Organizational Intelligence](https://hotfix.jobs/jobs/055a44bc-2ab2-4d24-bc52-ab326cc477fb) - Vanta - Remote - $176k – $207k/yr

**Apply:** https://hotfix.jobs/jobs/fe9c24f0-6ba5-4d79-ac4e-88906ab2e779
**Canonical:** https://hotfix.jobs/jobs/fe9c24f0-6ba5-4d79-ac4e-88906ab2e779