2 Domain Adaptation Jobs

Vimaan is looking to onboard multiple Machine Learning Engineers in Bengaluru, India to drive the development of computer vision and machine learning algorithms to power our cutting-edge wall-to-wall warehouse inventory tracking and verification platform. This is a unique opportunity to exploit a treasure cove of unseen real-world data coming from a multi-camera perception system and develop large-scale computer vision and deep learning models to build a product that creates a disproportionate value for the warehouse industry. The role involves hands-on CV/ML software development and deployment from understanding the product requirements, defining Computer Vision functional specs to designing, developing, and deploying CV/ML models in production at scale. The ideal candidate for the Machine Learning Engineer position should have an MS in computer vision, machine learning, AI, applied mathematics, data science, or related technical fields, or a BS with 3+ years of experience in Computer Vision/Machine Learning. They should possess hands-on experience in developing new learning algorithms for computer vision tasks such as object detection, object tracking, instance segmentation, activity detection, depth estimation, optical flow, multi-view geometry, domain adaptation, adversarial and generative models, as well as representational learning with varying amounts of data. Knowledge of current deep learning literature and the mathematical foundations of machine learning is essential. Experience with popular object detection frameworks such as YOLO, SSD, or Faster R-CNN is considered a plus. The candidate should have the ability to train and debug deep learning systems, gain deep insights into data characteristics, and map those to appropriate model architectures. Experience working with inputs from multiple cameras and input modes is advantageous, as is experience in AI Infrastructure, Machine Learning Accelerators, On-Device Optimization, and training and deploying deep learning models on GPU-accelerated platforms. Strong programming skills with Python and experience with ML/DL frameworks like Tensorflow, Pytorch, etc., are required. Prior experience in deploying machine learning models in production environments, working with cloud platforms (e.g., AWS, Azure, Google Cloud), and familiarity with data pre-processing, augmentation, and visualization tools and libraries are necessary. The ideal candidate should be highly motivated, passionate, possess a strong work ethic, and be able to work effectively in a team or independently under supervision in a matrix management environment. Effective communication skills, problem-solving abilities, attention to detail, and a passion for staying at the forefront of technology advancements in machine learning and computer vision are key attributes. They should be able to work in a fast-paced, high-pressure startup environment, adapt to rapidly changing requirements, and convey complex technical concepts to non-technical stakeholders. The Machine Learning Engineer will be responsible for researching, designing, and developing machine learning algorithms and models for various tasks of detection, recognition, and classification for warehouse inventory management. They will implement and optimize deep learning architectures, explore techniques like transfer learning and data augmentation, guide annotation teams, curate and pre-process annotated datasets, collaborate with MLOps for integrating machine learning models into production systems, and conduct thorough performance analysis and evaluation of models using appropriate metrics and tools. It is essential for the Machine Learning Engineer to stay up-to-date with the latest advancements in machine learning and computer vision research and integrate relevant findings into the solutions developed at Vimaan. ABOUT VIMAAN Headquartered in Silicon Valley, with team members around the world, Vimaan is comprised of computer vision and hardware technologists, as well as warehousing domain experts with a successful history in technology startups. Vimaan's primary mission is to deliver computer vision and machine learning solutions to solve long-standing inventory visibility, accuracy, and quality challenges in the supply chain.,

Job Title: R&D Lead Location: India Employment Type: Full-time Department: EXL AI Innovation Reports To: Salary Range: About Us EXL (NASDAQ: EXLS) is a $7 billion public-listed NASDAQ company and a rapidly expanding global digital data-led AI transformation solutions company with double digit growth. EXL Digital unit spearheads the development and implementation of Generative AI (GenAI) business solutions for our clients in Banking & Finance, Insurance, and Healthcare. EXL has partnered with NVIDIA AI Foundry As a global leader in analytics, digital transformation, and AI innovation, EXL is committed to helping clients unlock the potential of generative AI to drive growth, efficiency, and innovation. Job Description: If you are a smart, self-motivated Machine Learning Scientist with a passion for advancing the field of Generative AI, we have an excellent opportunity for you. We are seeking candidates with deep expertise in developing and fine-tuning LLMs, RAG, agentic solutions, and knowledge graph technologies to drive innovative solutions in GenAI. You will be at the forefront of pioneering advancements in AI, working alongside some of the brightest minds in an exciting R&D environment to build cutting-edge capabilities that redefine the future of artificial intelligence. Job Responsibilities: Develope initiatives in the GenAI domain, focusing on cutting-edge technologies like Large Language Models, Retrieval-Augmented Generation, and autonomous agents. Design and implement advanced workflows for integrating LLMs into real-world applications across various domains such as Finance, Insurance, and Healthcare. Develop and fine-tune domain-specific LLMs to optimize performance, using techniques like prompt engineering, adapter-based tuning, or low-rank adaptation. Drive the development of retrieval-augmented systems by combining LLMs with document retrieval, clustering, and search techniques. Stay at the forefront of AI advancements by reading, adapting, and implementing cutting-edge research to solve real-world challenges. Document research findings, methodologies, and implementations for internal and external stakeholders. Qualifications: Experience: 2-5 years in AI/ML research and development, with at least 1-2 years focusing on GenAI, LLMs, or related fields. Education: Master's or PhD in Computer Science, AI, or a related field from a top-tier institution is highly preferred. Required Skills: Core Expertise: Proven experience with Large Language Models (e.g., GPT-4, BERT, LLaMA, PaLM) and fine-tuning them for domain-specific applications. In-depth knowledge of Retrieval-Augmented Generation workflows, including retrieval system design and document store integration. Hands-on experience developing and deploying autonomous agents for complex problem-solving tasks. Tools & Frameworks: Proficiency in deep learning frameworks like TensorFlow, PyTorch, or Hugging Face Transformers. Experience with distributed training and optimization on GPUs and TPUs. Familiarity with cloud ecosystems (AWS, Azure, Google Cloud) practices for scalable deployment. Research & Development: Ability to read and adapt cutting-edge research papers for applied solutions in LLMs and knowledge graphs. Expertise in domain adaptation, few-shot learning, and zero-shot reasoning. Strong understanding of generative models, including VAEs, GANs, or diffusion models, and their integration with LLMs. Problem Solving: Demonstrated ability to address challenges in unstructured data processing, including NLP and multimodal scenarios. Experience with document retrieval, clustering, and unsupervised learning techniques. Preferred Skills: Experience with LLM fine-tuning and building Agentic systems for domain LLMs. Experience with reinforcement learning and fine-tuning via RLHF (Reinforcement Learning with Human Feedback). Knowledge of large-scale optimization methods for model training and inference. Familiarity with knowledge distillation and efficient model compression techniques. Strong collaboration and communication skills, with a proven ability to lead teams. Experience with MoE based architecture and knowledge on federated learning.