What You'll Do

Lead research efforts in machine learning for agentic flows, with a focus on speech/audio understanding and decision-making.
Explore, design, and evaluate state-of-the-art methods, running rigorous benchmarks and experiments.
Translate research outcomes into robust, production-ready solutions in collaboration with engineering and product teams.
Define metrics, measurements, and error analysis frameworks to ensure robustness and reliability in production.
Monitor deployed models, analyze performance regressions, and drive continuous improvements in accuracy and efficiency.
Act as the team’s primary research authority, staying up to date with the latest advances in ML, deep learning, and speech/audio processing, and applying them to our platform.

What We’re Looking For

5–7+ years of hands-on experience in data science and applied machine learning.
Strong background in both classical ML techniques and modern deep learning methods.
Proven experience deploying, monitoring, and iterating on models in production.
Experience designing and executing rigorous experiments, benchmarks, and error analyses.
Familiarity with agentic or sequential-flow systems (multi-step decision making, fallback logic, orchestration).
Proficiency in Python and ML/DL frameworks (e.g. PyTorch, TensorFlow, scikit-learn).
Strong problem-solving skills, curiosity, and ability to learn quickly.
Great collaborator who values ownership, autonomy, and teamwork.
Experience working in a fast-growing startup, with the ability to operate in a dynamic, fast-paced environment.
Bonus: prior work with speech/audio signals (e.g. ASR, embeddings, signal processing).