Top AI & Machine Learning Jobs

Lead Datadog AI Research’s post-training and autonomous-agent research team. Set technical direction across reinforcement learning, simulated environments, synthetic data generation, evaluation, and agent training. Mentor researchers and research engineers while remaining hands-on in experimentation, model development, and technical problem-solving. Collaborate with Product and Engineering to commercialize research advances, contribute to publications and conferences, and open-source model artifacts and benchmarks.

Leads hands-on research in foundation models, world models, and multimodal learning for distributed-system telemetry. Owns research from problem framing through large-scale model training and evaluation, advances pre-training and representation learning, mentors researchers and engineers, sets technical direction, collaborates with product and engineering teams, and contributes to publications, conferences, open-source artifacts, and benchmarks.

Leads hands-on research in post-training and autonomous agents, including reinforcement learning, simulated environments, synthetic data, training loops, and evaluation. Owns research from problem framing through experimentation and model development, sets technical direction, mentors researchers, collaborates with product and engineering teams, and contributes to publications, conferences, open-source artifacts, and benchmarks.

Leads an eight-person anomaly detection engineering team, setting technical direction, owning roadmap execution, and delivering scalable signal-finding systems. Partners with product managers and applied scientists to evaluate detection quality. Builds and operates traditional machine learning and LLM systems in production, drives AI-native development practices, and establishes standards for safety, validation, ownership, and execution.

Lead a team conducting foundational research in foundation models, world models, and multimodal learning for observability and security data. Set technical direction, guide large-scale model training and evaluation, mentor researchers and research engineers, and collaborate with product and engineering teams to commercialize research. Contribute to publications, conferences, open-source artifacts, and benchmarks while remaining hands-on in research strategy, experimentation, and model development.

As an AI Researcher at Datadog, you'll conduct cutting-edge research in AI, focusing on foundation models and AI agents for observability and reliability, collaborating with cross-functional teams, and contributing to scientific publications.

Leads an AI-native engineering team developing production LLM and agent software for code intelligence. Owns strategy and execution across source code indexing, pull-request agents, code-writing agents, and closed-loop evaluation systems. Sets technical direction, partners with customers and product teams, and establishes standards for safety, validation, agent autonomy, and product delivery.

Conduct an end-to-end AI research project with Datadog AI Research and Carnegie Mellon University. The intern will frame problems, train and evaluate models using GPUs and large-scale data, run experiments, collaborate with DAIR mentors and a faculty advisor, publish findings, and potentially transition successful research into production systems.

Applied AI Interns will research and benchmark AI models, develop scalable generative AI and machine learning features, and deploy, monitor, and maintain production AI systems. They will analyze large-scale observability data, investigate emerging model architectures, present research, and collaborate with engineering and product teams to deliver customer-facing capabilities.

Own the applied science direction for GenSim, Datadog’s generative simulation environment for training and evaluating AI agents. Define methodology and quality metrics for post-training data, improve simulation realism and failure difficulty, and develop scalable production systems integrated into training loops. Lead LLM and agent evaluation approaches, determine how synthetic data should be applied, and collaborate across engineering and applied science teams.