About the job
This job will design, develop, and implement machine learning models and algorithms to solve complex problems. You will work closely with data scientists, software engineers, and product teams to enhance services through innovative AI/ML solutions. Your role will involve building scalable ML pipelines, ensuring data quality, and deploying models into production environments to drive business insights and improve customer experiences.
Responsibilities
Develop and optimize machine learning models for various applications.
Preprocess and analyze large datasets to extract meaningful insights.
Deploy ML solutions into production environments using appropriate tools and frameworks.
Collaborate with cross-functional teams to integrate ML models into products and services.
Monitor and evaluate the performance of deployed models.
Qualifications
Minimum
3+ years relevant experience and a Bachelor’s degree OR Any equivalent combination of education and experience.
Experience with ML frameworks like TensorFlow, PyTorch, or scikit-learn.
Familiarity with cloud platforms (AWS, Azure, GCP) and tools for data processing and model deployment.
Several years of experience in designing, implementing, and deploying machine learning models.
Preferred
PhD in Computer Science, AI/ML, NLP, or a related field — with research in one or more of: LLM agents, multi-agent systems, tool-use, reasoning, planning, dialogue systems, or reinforcement learning
Strong engineering skills — you can take a research idea from paper to production. Python is second nature. You've built systems, not just run experiments.
Deep familiarity with agentic frameworks and architectures — LangChain, Google ADK, or custom orchestration systems. You understand the tradeoffs.
Experience with LLM APIs and tool-use patterns — function calling, structured outputs, retrieval-augmented generation, chain-of-thought, and prompt engineering at scale
Understanding of evaluation methodology for AI systems — how to measure agent performance beyond academic benchmarks, including safety, hallucination rates, and task completion in adversarial conditions
Published research is a plus — but shipping code matters more than citation count