Senior Machine Learning Engineer - Dispatch, Surge & Incentives

About Rafeeq

Rafeeq is a rapidly growing on-demand delivery platform connecting customers with restaurants and couriers across the region. We're solving complex marketplace problems at scale - balancing supply and demand in real-time to create exceptional experiences for everyone on our platform.

The Role

You will focus on three interconnected problems that are critical to our business:

1. Dispatch Optimization - Intelligent courier assignment, order batching, and routing

2. Surge Pricing - Dynamic pricing to balance supply and demand in real-time

3. Incentive Systems - Smart bonus zones and payments to position couriers where they're needed

These systems directly impact courier earnings, customer wait times, and our marketplace efficiency. Your models will be making thousands of decisions per minute in production.

What You'll Do

Dispatch Optimization

• Build ML models for optimal courier-to-order assignment considering distance, courier state, acceptance probability, and order characteristics

• Implement order batching algorithms to allow couriers to deliver multiple orders efficiently

• Research and prototype advanced techniques: Graph Neural Networks, combinatorial optimization

• Optimize for multiple objectives: delivery time, courier earnings, customer satisfaction, platform efficiency

Dynamic Pricing (Surge)

• Design and deploy surge pricing models that respond to real-time supply-demand imbalances

• Build demand forecasting models at geographic zone and hourly/sub-hourly granularity

• Incorporate external factors: weather, events, seasonality, holidays

• Run A/B experiments to optimize pricing strategies for both customer experience and marketplace balance

Incentive Systems

• Develop models to predict where courier supply will be needed 30-60 minutes in advance

• Build intelligent bonus zone systems to proactively position couriers

• Design incentive structures that maximize courier earnings while improving platform efficiency

• Create attribution models to measure incentive effectiveness

Core Responsibilities

• Model Development: Research, prototype, and deploy ML models for dispatch, pricing, and incentives

• Feature Engineering: Build real-time feature pipelines using geospatial, temporal, and marketplace data

• Production Systems: Deploy models in high-throughput, low-latency environments (p99 < 100ms for dispatch)

• Experimentation: Design and analyze A/B tests to measure impact on key metrics (ETA, courier earnings, order volume)

• Collaboration: Work closely with Product, Engineering, and the ML team to ship features end-to-end

• Monitoring: Build dashboards and alerts to track model performance and marketplace health

What We're Looking For

Required:

• 5+ years of experience in ML/Data Science with at least 3+ years deploying models to production

• Strong ML fundamentals: Regression, classification, time-series forecasting, optimization

• Expert Python skills and deep experience with ML libraries (Scikit-learn, XGBoost)

• Advanced SQL for complex feature engineering and data analysis

• Production ML experience: Real-time inference, model serving, monitoring, A/B testing

• Geospatial data experience: Working with lat/lon, distance calculations, zone-based aggregations

• Bachelor's or Master's degree in Computer Science, Statistics, Mathematics, Operations Research, or related field

Strongly Preferred:

• Experience in marketplace companies (food delivery, ride-hailing, e-commerce)

• Domain expertise in dispatch, routing, dynamic pricing, or incentive systems

• Experience with optimization algorithms (linear programming, reinforcement learning, combinatorial optimization)

• Familiarity with streaming data (Kafka, Kinesis) and real-time ML systems

• Experience with MLOps tools and practices (feature stores, model registries, monitoring)

• Knowledge of geospatial libraries (PostGIS, GeoPandas, H3, Kepler.gl)

• Publications or contributions to open-source ML projects

Nice to Have:

• Experience with Graph Neural Networks (GNNs) for routing/dispatch problems

• Background in operations research or supply chain optimization

• Experience with causal inference for measuring treatment effects

• PhD in a relevant field

Why Join Rafeeq?

🎯 High Impact: Solve critical marketplace problems affecting thousands of couriers and millions of customers daily

🚀 Greenfield Opportunity: Build these systems from scratch with modern tools and best practices

🌍 International Team: Remote-first culture, work from anywhere, competitive salary in USD/EUR

⚡ Fast-Paced: Ship quickly, iterate based on data, see immediate impact of your work

🧠 Technical Excellence: Work with a strong ML team led by an experienced Team Lead, learn from each other

💰 Competitive Compensation: Market-rate salary in foreign currency + equity in a growing company

📊 Ownership: Own entire problem domains end-to-end, from research to production to iteration

Problems We're Solving

Food Delivery Focus: Our primary challenge is optimizing the food delivery marketplace with focus on ETA accuracy, intelligent dispatch, and dynamic pricing based on demand forecasts.

Taxi Vertical (Future): We're also building a taxi service facing incentive optimization challenges - demand prediction, surge pricing, and smart courier positioning.

You'll initially work on food delivery problems, with potential to expand to taxi as we scale the team.

Hiring Process

1. Application Review: HR team screens applications - we respond quickly with initial interest

2. Team Interview: Technical discussion with ML team members about your experience and approach

3. Final Interview: Deep technical conversation with Anton (Team Lead) covering ML expertise, problem-solving, and collaboration

4. Offer: Fast decision-making, typically 2-3 weeks from application to offer

We hire internationally and move faster than most companies - strong candidates often receive offers within 2-3 weeks.

What Success Looks Like - First 90 Days

• Deep understanding of our dispatch, pricing, and incentive systems - current state and opportunities

• Ship first model improvement to production (dispatch OR surge OR incentives)

• Design and launch A/B experiment to measure impact

• Build monitoring dashboards for your problem area

• Establish strong working relationships with Product, Engineering, and ML team

• Contribute to technical roadmap for your focus area