Research Scientist Intern - Post-Training (RLHF) at Jasper

France

Apply Now

Not SpecifiedCompensation

InternshipExperience Level

InternshipJob Type

UnknownVisa

Artificial Intelligence, TechnologyIndustries

Requirements

Currently enrolled in a Ph.D. or M.Sc. program in Machine Learning, applied mathematics, or computer science (Ph.D. preferred)
Experience with RLHF techniques applied to either Large Language Models or Diffusion Models
Genuine interest in the field and strong motivation to contribute to open-source initiatives, with proven track record through personal projects or previous experience in deep learning, especially generative models (e.g., diffusion models, GANs, VAEs, transformers)
Strong coding abilities in Python and deep learning frameworks (PyTorch, TensorFlow, JAX)
Strong critical thinking and problem-solving mindset, coupled with excellent teamwork skills
Available for a period of 6 months
Located in Paris (hybrid setup)
Nice to have
Strong programming Python skills, including software engineering best practices to produce high-quality code
Experience with distributed training and large-scale systems on GPU clusters
Experience with large-scale data processing
Contributions to open-source projects
Proven track record of achieving significant results as demonstrated by first-authored publications in major conferences and journals such as CVPR, ECCV, ICCV, ICLR, NeurIPS

Responsibilities

Conduct literature reviews, propose and implement innovative methods to fine-tune text-to-image models with reinforcement learning methods (RLHF, DPO)
Participate in the fine-tuning stages of the training of large-scale text-to-image models, conduct rigorous ablations, design evaluation metrics, and analyze model performance
Document research findings, prepare technical reports, and participate in the external communication of the results
Contribute to an ambitious open-source project, publish research findings, and engage with the broader AI community

Skills

Key technologies and capabilities for this role

RLHFDiffusion ModelsText-to-ImageImage EditingModel Fine-TuningMachine LearningComputer VisionNatural Language ProcessingOpen Source Models

Questions & Answers

Common questions about this position

What is the location and work arrangement for this internship?

The role is open to candidates located in Paris and follows a hybrid setup, requiring you to come into the office when necessary. Jasper is a remote-first organization overall.

What are the required qualifications for this Research Scientist Intern position?

Candidates must be currently enrolled in a Ph.D. or M.Sc. program in Machine Learning, applied mathematics, or computer science (Ph.D. preferred), with experience in RLHF techniques applied to Large Language Models or Diffusion Models, and a genuine interest in the field.

How long is the internship duration?

The internship duration is 6 months.

What will I be working on as a Research Scientist Intern?

You will design RLHF techniques for diffusion models, conduct literature reviews, implement fine-tuning methods like RLHF and DPO, participate in model training and evaluation, document findings, and contribute to open-source projects.

What is the compensation for this internship?

This information is not specified in the job description.

Jasper

AI tool for efficient content creation

About Jasper

Jasper.ai provides an AI-powered tool designed to assist in content creation, specifically for producing high-quality blogs quickly and cost-effectively. This tool helps marketers, especially in the B2B SaaS sector, to generate effective campaigns by streamlining the writing process. Jasper.ai's service operates on a business model that enhances productivity for clients who need a large volume of content, such as marketing agencies and e-commerce businesses. A key differentiator for Jasper.ai is its commitment to ethical practices; it ensures that its AI models are not trained on client data, maintaining user privacy and confidentiality. Additionally, the company does not retain ownership of the content generated through its tool, and all data is stored securely in U.S.-based data centers. The goal of Jasper.ai is to provide a reliable and efficient content creation solution while upholding high standards of data security.

Austin, TexasHeadquarters

2020Year Founded

$121.6MTotal Funding

SERIES_ACompany Stage

Enterprise Software, AI & Machine LearningIndustries

501-1,000Employees

Benefits

Competitive salary

Unlimited vacation policy

401(k) matching

Health, dental, & vision benefits

Happy hour & fun events

Paid holiday & sick days

Risks

Increased competition from OpenAI's ChatGPT integration with iPhone may attract Jasper's market.

Integration challenges from Clipdrop acquisition could dilute Jasper's text-focused content creation.

Market saturation in generative AI could challenge Jasper's unique value proposition.

Differentiation

Jasper offers a unique AI knowledge layer for marketing with Brand IQ and Marketing IQ.

The company ensures data security by not training AI models on client data.

Jasper's AI tool enhances productivity by reducing writing costs for high-quality content.

Upsides

Jasper's acquisition of Clipdrop expands its services to include image creation and editing.

The AI Image Suite targets B2C marketers, improving workflows and campaign launches at scale.

Generative AI's growing demand offers Jasper opportunities for personalized marketing campaigns.

Land your dream remote job 3x faster with AI

Try Jobo Free