About the job
Join Ema as a Principal Machine Learning Engineer
Ema is pioneering the future of AI technology, enabling every employee within enterprises to unlock their full creative and productive potential. Our cutting-edge platform allows businesses to assign repetitive tasks to Ema, the Universal AI Employee—a robust, secure, and intelligent collaborator that seamlessly integrates across various workflows and systems.
Founded by seasoned professionals from Google, Coinbase, and Okta, and supported by prominent investors and angels, Ema is headquartered in Silicon Valley and Bangalore. We operate as a hybrid team, and we expect our team members to work from the office three days a week.
Role Overview & Key Responsibilities
This leadership role is pivotal and encompasses architecture, execution, and organizational development, driving the direction of our AI and ML initiatives at Ema. We are in search of a technical leader in AI and ML who can translate vision into reality. As a Principal ML Engineer, you will be instrumental in shaping the machine learning roadmap, designing large-scale ML systems, fostering innovation, and ensuring the accuracy and performance of our diverse models (including LLM, SLM, and Custom Models) at scale. You will work collaboratively across teams—ranging from research and product to infrastructure and data—while mentoring senior engineers and influencing strategic and executional decisions at the company-wide level.
Responsibilities
- Guide the technical strategy for GenAI and agentic ML systems that support enterprise-grade AI agents, covering reasoning, retrieval, tool utilization, and integration with various SaaS products.
- Design and implement scalable production pipelines for model training, fine-tuning, retrieval (RAG), agent orchestration, and evaluation, ensuring robustness, low latency, and continuous learning.
- Develop and manage a multi-year ML roadmap for GenAI infrastructure, including agent frameworks, RAG systems, evaluation loops, and integration with MCP, browser, and vision pipelines.
- Incorporate cutting-edge ML methodologies and research (such as deep learning, large models, recommender systems, and LLMs) into Ema’s products and infrastructure.
- Experiment with and integrate advanced ML and LLM innovations (e.g., reasoning, memory architectures, multi-modal perception, long-context models, and autonomous agents) into the platform.
- Balance trade-offs among accuracy, latency, cost, interpretability, and real-world reliability throughout the agent lifecycle, from prompt design to orchestration and execution.
