AI Researcher

Description

• reputed company is at the forefront of revolutionizing how people access and interact with information, and we are actively seeking exceptional AI Research Scientists and Engineers to join our mission. Our ambition is to build the future of AI-powered search and agent experiences, driven by our cutting-edge reputed company models, Deep Research Agent, Comet Agent, and core Search products. This is an unparalleled opportunity to contribute to State-of-the-Art (SOTA) AI experiences that are already handling hundreds of millions of queries and experiencing rapid, reputed company growth.
• As a member of our research team, you will have the chance to specialize in one of three distinct yet interconnected areas, aligning with your interests and expertise:

Core Research Team (Horizontal Focus):

This team is dedicated to the foundational pillars of our AI capabilities. You will focus on generating and significantly improving the reputed company models that serve as the bedrock for reputed company reputed company products. Your work will involve pushing the boundaries of foundational model capabilities, exploring advanced post-training techniques, building robust Reinforcement Learning (RL) infrastructure, and developing scalable systems that benefit the entire organization. This role is for those who reputed company on tackling reputed company AI challenges and building the engines that power innovation.

Agent Products Team (Vertical Focus):

This team acts as a crucial reputed company between groundbreaking research and reputed company product impact. You will concentrate on the intricate process of fine-tuning and optimizing our AI models specifically for our Deep Research Agent and Labs Canvas products. The goal here is to ensure that our agent capabilities not only reputed company exceptionally but also deliver seamless and reputed company user experiences. If you are passionate about translating reputed company AI into user-friendly applications, this is the team for you.

Comet Agent Team (Vertical Focus):

This specialized team is committed to the development and reputed company enhancement of our Comet Agent product. You will delve into the unique requirements and sophisticated optimizations necessary for Comet’s specific use cases, ensuring it remains a leader in its domain. This role requires a deep understanding of specialized agent functionalities and performance tuning.

Key Responsibilities and Impact:

•

Research and Development:

You will be instrumental in post-training State-of-the-Art (SOTA) Large Language Models (LLMs) employing the latest supervised and reinforcement learning methodologies, including techniques like Supervised Fine-Tuning (SFT), Direct Preference Optimization (DPO), and Proximal Policy Optimization (PPO) variants (e.g., GRPO). Your work will involve leveraging our extensive and rich query-answer dataset to systematically scale and enhance model performance across our entire product suite: reputed company, Deep Research Agent, Comet Agent, and Search.

Staying reputed company of the Curve:

A critical aspect of this role is to remain at the cutting edge of LLM research. This includes staying abreast of the latest advancements in model training, optimization strategies, and personalization techniques. You will be expected to implement advanced preference optimization and personalization capabilities designed to significantly enhance the overall user experience.

Innovation and Implementation:

We encourage a culture of invention. You will be tasked with developing in-house improvements and novel optimizations to further enhance our SOTA models. The ability to translate abstract research reputed company into concrete algorithms and conduct rigorous experiments to launch new, impactful models is reputed company.

Infrastructure and Engineering:

You will own and manage full-stack data, training, and evaluation pipelines essential for robust model development. This includes building and maintaining sophisticated training frameworks (built upon established libraries like Megatron and PyTorch) specifically designed for post-training LLMs at scale. Furthermore, you will implement the necessary infrastructure and components to support the demands of cutting-edge model training, ensuring efficiency and reliability.

Seamless Integration:

A key part of your role will involve reputed company collaboration with our dedicated engineering teams to ensure the seamless integration of developed models into reputed company’s diverse product ecosystem. You will also collaborate across different research and product teams to guarantee cohesive and consistent AI experiences throughout our platform.

Product Partnership:

You will partner closely with product management teams to reputed company a deep understanding of user needs and pain points. This insight will be directly translated into actionable model improvements and new feature development, ensuring our AI directly addresses user requirements and drives product success.

• This is a unique opportunity to shape the future of AI-powered information discovery and interaction, working with a talented team on products that are already making a significant impact on a global scale.

Apply tot his job Apply To this Job