This role is for one of the Weekday's clients

Salary range: Rs 1500000 - Rs 3000000 (ie INR 15 - 30 LPA)

Min Experience: 3+ years

Location: Bangalore, Karnataka
JobType: full-time

We are looking for an AI Research Engineer to work on advanced Computer Vision, Generative AI, Vision-Language Models (VLMs), and Vision-Language-Action (VLA) systems for autonomous robotics.

The role combines AI research with production engineering taking ideas from research and experimentation through evaluation, optimization, and deployment on real-world robotic systems.

This is a research-focused AI role working on Computer Vision, Generative AI, and multimodal models for autonomous systems. The role involves developing Diffusion Models, VLMs, and data-centric AI solutions to improve machine perception and understanding, while taking research models through experimentation, optimization, and deployment on real-world edge systems.

Requirements

Key Responsibilities:

• Research and develop Diffusion-based Generative AI models for synthetic data generation, defect simulation, photorealistic environments, and domain adaptation.
   
• Design and train VLMs/VLAs connecting text instructions, visual data, CAD/spatial information, and sensor inputs for scene understanding and intelligent decision-making.
   
• Build scalable auto-annotation and data-centric AI pipelines using Active Learning, Self-Training, Pseudo-Labeling, Weak Supervision, and Synthetic Data.
   
• Develop AI workflows capable of handling millions of images, video frames, and point clouds with minimal manual annotation.
   
• Optimize models for edge deployment using INT8 quantization, LoRA, Knowledge Distillation, TensorRT, ONNX Runtime, CUDA, and C++.
   
• Work with NVIDIA Jetson-class edge hardware and integrate AI models with robotics systems.
   
• Own the complete research lifecycle: problem definition → literature review → prototyping → training → evaluation → optimization → production handoff.
   
• Collaborate with Computer Vision, Perception, Robotics, and Controls teams.
   
• Contribute to technical documentation, research publications, and mentor junior engineers/interns.

Requirements:

• 3 to 7 years in Deep Learning, AI Research, Computer Vision, or related R&D; M.S./Ph.D. in CS, EE, Robotics, or related fields is highly relevant.
   
• Strong hands-on experience with Diffusion Models: DDPM, LDM, ControlNet, Generative Modeling, Synthetic Data, or Domain Adaptation.
   
• Strong experience with VLMs / Multimodal Transformers, such as CLIP, BLIP-2, LLaVA, Flamingo, or equivalent architectures.
   
• Proven experience with Active Learning, Pseudo-Labeling, Self-Training, Weak Supervision, and automated annotation.
   
• Advanced Python + PyTorch skills; JAX is a plus.
   
• Experience with scalable training frameworks such as PyTorch Lightning, DeepSpeed, Ray, or equivalent.
   
• Strong understanding of Probability, Optimization, Linear Algebra, and Information Theory.
   
• Ability to translate research concepts into production-ready AI systems.

Good to Have

• Robotics / Autonomous Systems / Embodied AI / Perception
   
• ROS 2, Isaac Sim, Open3D, Nav2, MoveIt 2
   
• 3D Computer Vision / Point Clouds
   
• TensorRT / ONNX Runtime / CUDA / C++
   
• Experience deploying models on NVIDIA Jetson or edge AI devices
   
• Strong research publications in AI/ML, Computer Vision, Generative AI, or Robotics.

Must-have skills

Computer Vision, Diffusion Models, AI Research

Good-to-have skills

Python, Pytorch, Vision Language Model