Job Title: Software Development Engineer, Trustworthy AI Company Name: Vijil, Inc Job Details: $150k-$200k/yrRemoteFull,Time Job Url: https://hiring.cafe/viewjob/l9nwlef1yavsmhsl Job Description: Posted 15h agoSoftware Development Engineer, Trustworthy AI@ Vijil, IncView All JobsUnited States$150k-$200k/yrRemoteFull TimeResponsibilities:solve problems, detect vulnerabilities, assess harmsRequirements Summary:Senior software engineer with 8+ years exp in SDE roles; experience in LLM training, tuning, NLP models, and ML frameworks; strong Python/Java/C++ skills; collaboration with researchers; security and safety focus for AI systems.Technical Tools Mentioned:TensorFlow, PyTorch, JAX, Python, Java, C++, Git Vijil is looking for a Senior Software Engineer to develop methods for improving the security and safety of large language models. Description About UsVijil is a venture-funded AI software startup on a mission to help organizations build and operate intelligent agents that people can trust. We are building tools and services to enable AI developers to continuously improve the security and safety of custom large language models.About the RoleAs a software developer at Vijil, you will build upon the latest research methods in trustworthy ML and generative AI to deliver cloud services to AI developers who are customizing LLMs with enterprise-specific data. You will collaborate with applied scientists and technologists to build capabilities for testing, tuning, and deploying LLMs with security and safety. Your day-to-day responsibilities include:Working closely with applied scientists, academic collaborators, and senior leaders, and prospective users to solve problems facing enterprises that want to deploy LLMs in productionImplementing methods to detect LLM vulnerabilities to attacks such as prompt injection, PII disclosure, model evasion, jailbreaking, and model extraction.Implementing methods to assess LLM propensities for harm such as toxicity, unfair, and unethical behaviorDesigning and developing new techniques to reduce vulnerabilities to attack and mitigate propensities for harmsAbout YouYou take ownership of your customer's problem, dive deep into root causes, invent new methods anchored in CS/AI/ML first principles to address the problem, deliver rapid results, and continuously enhance the customer's experience. As part of a highly collaborative team, you rely on your exceptional (written and verbal) communication skills to inform and persuade. You have a strong analytical mind, you approach challenges with a problem-solving stance, and you have the grit and resilience to deal with setbacks. You find this work meaningful because you know deeply that we need to build trust into AI/ML systems from the inside-out in order to realize their transformative potential. Minimum QualificationsMS or Ph D in Computer Science, Applied Math, or equivalent discipline8+ years industry experience as an SDE or equivalent job2+ years experience in systems for LLM training and tuning.4+ years experience with training or inference of NLP models.4+ years experience with ML frameworks: TensorFlow, PyTorch, JAX.Demonstrable proficiency in programming with Python, Java, or C++ (GitHub link appreciated)Proven competence in software engineering practices (version control, unit testing, documentation, code review, deployment, operations, on-call rotation, customer engagement)Proven ability to work collaboratively with research scientistsDemonstrable written and verbal communication skills (writing samples appreciated)Preferred QualificationsMS or Ph D in Machine Learning, NLU, Neural Networks, or equivalentExperience with building LLMs or agents.Familiarity with adversarial ML, LLM red-teaming, or responsible AIExperience working with cloud services on AWS, Azure, or GCPBenefitsCompetitive base pay and equity.Health, dental, and vision insurance plans.Flexible time-off policy.Paid parental and family leave.Immigration sponsorship.How to ApplyIf you are excited about making a meaningful impact on mechanisms to improve the security of generative AI models, please submit your resume and any examples of your relevant work. For any questions, email us at [email protected].Vijil is an equal opportunity employer and complies with all applicable federal, state, and local laws pertaining to fair employment practices. About the Company Vijil is an AI software startup on a mission to help organizations build and operate autonomous agents that they can trust. Founded by technologists who built the deep learning platform in Amazon SageMaker at AWS and established responsible AI practices at Twitch and Splunk, Vijil is helping AI developers build-in trust into open-source large language models. If you're passionate about improving the explainability, security, and safety of neural networks that will be used in business-critical applications over the next decade, you'll enjoy making an impact at Vijil. Join us today to grow with the company and its products and customers.