Applied Scientist jobs in Tel Aviv-Yafo

Overview Microsoft Teams is the hub for teamwork that integrates all the people, content, and tools your team needs to be more engaged and effective. It is core to Microsoft’s modern work, modern life & modern education value prop. We are reinventing the way people communicate and work together across the globe.   We are looking to hire a PhD (or published MSc) candidate for a 12-week, full-time internship to join CMD Labs – an applied science team within Microsoft Teams – to work on improving collaborative agents by applying existing or novel research. Our flagship AI application for Teams is the Copilot agent that works in multi-party collaborative settings (meetings, group chats, channels) and enables groups of humans and agents to more efficiently achieve their shared goals. While most human/agent interactions are textual, within live Teams meetings the agent also interacts in real time via speech. Unlike personal agents (both audio and text), multi-party collaborative agents are faced with significant challenges in implicit context understanding, proactivity, and conversational capabilities. The intern will be onboarded to our current research directions, and to our non-confidential datasets and related code, and work on improving the existing algorithms as well as proposing novel solutions, in any of the following topics: • Training/fine-tuning speech-enabled models (e.g. Qwen-omni-8) to support speech in multi-party settings such as business meetings. • Improve and devise new methodologies for evaluation of audio/text collaborative agents. • Improve and devise better RL recipes for training/fine-tuning collaborative models including synthetic data generation, and the development and calibration of graders. Given substantial scientific novelty of the approach and results, collaboration on a mutual publication is encouraged. Responsibilities • Conduct experiments, create and validate metrics, and develop candidate algorithms to improve the existing capabilities of current models and agent implementations in any of the above topics. • Collaborate closely with CMD Labs researchers and engineers to leverage existing assets, datasets, and ensure results can contribute to the product. • Embody our culture and values. Qualifications Required • Currently enrolled in a PhD program (or published candidate in MSc program) in Computer Science, Electrical or Computer Engineering, Statistics, or a related field. • Practical experience in training, fine-tuning, and prompt engineering of transformer models or LLMs. • Practical Python coding experience leveraging modern coding agents. Preferred     • Field of research and publications directly related to Audio LLMs or collaborative models and agents. *** Please note that this is a 12 weeks full time intenship ****** This position will be open for a minimum of 5 days, with applications accepted on an ongoing basis until the position is filled. Microsoft is an equal opportunity employer. All qualified applicants will receive consideration for employment without regard to age, ancestry, citizenship, color, family or medical care leave, gender identity or expression, genetic information, immigration status, marital status, medical condition, national origin, physical or mental disability, political affiliation, protected veteran or military status, race, ethnicity, religion, sex (including pregnancy), sexual orientation, or any other characteristic protected by applicable local laws, regulations and ordinances. If you need assistance with religious accommodations and/or a reasonable accommodation due to a disability during the application process, read more about requesting accommodations.

Description About Us: Zenity is the leader in AI Agent Security and the first company to bring an agent-centric security platform to market. As enterprises accelerate AI agent adoption, we are establishing the security framework for how AI agents are secured and governed at enterprise scale. We deliver full-lifecycle visibility, governance, detection, prevention, and response for AI agents from build time to runtime, across SaaS, home-grown platforms, and end-user devices. Backed by $180M+ in total funding, including a $125M Series C led by Norwest, with participation from SoftBank Vision Fund 2 and Microsoft's M12, Zenity is trusted by Fortune 500 and Global 2000 enterprises worldwide. Join us in shaping how AI agents are secured at enterprise scale. About the Role: We're looking for a Senior Applied AI Scientist to sit at the frontier of AI security - turning emerging threats into the detection models that protect how AI is used inside the world's largest organizations. You'll be part of the AI Security Research department, working hand-in-hand with security researchers to translate threat intelligence into trainable signals that catch malicious behavior and security risks across the AI-powered workflows of Fortune 500 companies. You'll bring deep technical versatility - reaching for classical ML, deep learning, or agentic based approaches based on what the problem demands, and the evaluation rigor to know when a model is truly ready for the real world. If you want to define what AI security engineering looks like, not just practice it, this role is for you. What You’ll Do Responsibilities: • Build, train, and ship detection models end-to-end, from raw data to production • Choose the right method for each problem - traditional ML, deep learning, fine-tuned LLMs, agents or heuristics - based on theoretical insights turned into practical results. • Partner with security researchers to turn security research outputs and domain expertise into detection capabilities • Own evaluation: design benchmarks, build labeled datasets, and define production standards • Monitor models in production across all paradigms - ML, deep learning, LLM-based, and agentic systems to track degradation and ensure reliability • Iterate fast, with a tight feedback loop between model performance and product outcomes Requirements Requirements: • 5 years of hands-on ML and deep learning experience, with a track record of shipping, debugging, and diagnosing models in production • Data-first mindset: you know how to define the right evaluation criteria for each model - before and after shipping, to ensure it delivers real quality and value in production • Hands-on experience building and deploying agentic AI systems to production • Proficiency in Python; experience with PyTorch, scikit-learn, HuggingFace, or equivalent • Practical, applied mindset - focused on the problem, success metrics and impact, not lab research. • • Background in security, trust & safety, or content moderation - an advantage