Senior Data Scientist / ML Engineer

We are building a next-generation, adaptive route planning engine designed for extreme logistics complexity. Unlike traditional static optimization models, our system evolves by learning from human expertise. We capture the manual adjustments made by dispatchers and transform that “implicit knowledge” into scalable algorithmic assets.

We are looking for a versatile Data Scientist who thrives at the intersection of Optimization (OR) and Machine Learning. You will own the full ML lifecycle—from deep-dive data analysis and sophisticated feature engineering to selecting and implementing the right architectures (from Decision Trees to Transformers/RL) to bridge the gap between automated planning and human intuition.

Details: 
Start: ASAP 
Experience: 4+ years 
Schedule: Full time remote 
Mode: Fist 1-2 months - part time with future allocation to full time mode.

Key Responsibilities
Human-in-the-Loop Learning: Develop models that analyze the delta between algorithmic outputs and manual human corrections to “learn” implicit constraints and preferences.
Complex Constraint Satisfaction: Work on Large-scale Vehicle Routing Problems (VRP) involving multi-layered constraints and complex clustering.
End-to-End ML Ownership: Lead the entire pipeline: sophisticated data normalization, advanced feature engineering, model selection, training, and evaluation.
Strategic Analytics: Implement analytical frameworks (e.g., Balanced Scorecard approaches) to quantify performance and align model behavior with business KPIs.
Algorithmic Research: Experiment with diverse approaches including Gradient Boosted Trees, Attention mechanisms, and potentially Reinforcement Learning to capture the nuances of logistics planning.

Required Qualifications
Experience: 4+ years of experience as a Data Scientist or ML Engineer, with a proven track record of taking models from research to production.
Optimization Expertise: Strong understanding of optimization problems (VRP/TSP), heuristics, and working with complex constraints. Familiarity with tools like Google OR-Tools is a major plus.
Machine Learning Depth: Solid grasp of supervised learning and a willingness to explore Imitation Learning or RL.
Feature Engineering Mastery: You believe that data preparation and creative feature engineering are the keys to a model’s success. Expert-level Python and SQL skills are a must.
Architectural Flexibility: Ability to match the right tool to the problem—whether it’s a simple Decision Tree for early-stage wins or a Transformer-based architecture for complex sequence modeling.
Analytical Mindset: Experience in defining and tracking complex metrics; ability to perform “smart analysis” to find patterns in human behavior.

Bonus Points
Experience in the logistics or supply chain domain.
Background in Geospatial data.
Experience with Cloud Infrastructure (AWS Bedrock, SageMaker, etc.).