📍 Location: Chennai
💼 Experience: 4+ Years
🕒 Employment Type: Full-time
We are seeking a Data Scientist with strong expertise in predictive modeling, reinforcement learning, and simulation-based analytics. The ideal candidate will design and develop advanced algorithms for trajectory forecasting, behavior prediction, and risk analysis in dynamic, real-time environments.
Predictive Modeling: Design and implement models for trajectory forecasting, traffic participant behavior, and crossing probability estimation.
Risk Assessment: Develop and maintain time-shifted risk prediction mechanisms using sliding time windows.
Reinforcement Learning: Build and train multi-agent reinforcement learning (MARL) frameworks to simulate cooperative and competitive behaviors.
Simulation Integration: Collaborate with simulation teams to incorporate ground-truth data and replayable scenarios for model training and validation.
Scoring & Evaluation:
Build scoring algorithms across multiple data dimensions based on severity and impact.
Evaluate model performance using metrics such as precision, recall, and event-level accuracy.
Collaboration: Partner with data engineers to design scalable feature pipelines and real-time streaming data inputs.
Experience: Minimum 3 years in applied data science, preferably in real-time analytics or simulation-based systems.
Technical Skills:
Strong proficiency in Python, NumPy, and Pandas.
Hands-on experience with deep learning frameworks such as PyTorch or TensorFlow.
Domain Knowledge:
Expertise in time-series analysis, Bayesian models, or probabilistic forecasting.
Understanding of reinforcement learning, particularly in multi-agent environments.
Familiarity with vehicle kinematics, trajectory forecasting, or intelligent transportation systems (ITS).
Soft Skills: Excellent analytical thinking, problem-solving ability, and collaboration skills in cross-functional environments.
Experience working with large-scale simulation platforms or real-world traffic datasets.
Familiarity with cloud environments such as AWS, GCP, or Azure.
Background in computer vision or sensor fusion is a plus.