MLE 2 – Spatiotemporal Intelligence

Location: Pune, India (On-site)

Type: Full-Time

Dispatch Network is building the most efficient last-mile network in India from the ground up using technology and AI-powered optimization to drive efficiency and earnings for delivery partners. We operate across food delivery, quick commerce, grocery, ecommerce, and pharma.

Dispatch Network is building an adaptive logistics intelligence platform a system that learns city dynamics in real time and optimizes how goods move through urban space. Were moving from pilot to national scale, and the AI layer you build here will shape how fleets behave across Indias densest delivery environments.

Were hiring a Machine Learning Engineer II to independently design, build, and deploy spatiotemporal and forecasting models at scale. This role owns significant model pipelines and works across engineering to build reliable, real-time ML systems for complex logistics environments.

Build and own forecasting and spatiotemporal models (Transformers, diffusion forecasters, GNNs)

Develop geospatial intelligence using H3 indexing, mobility modeling, or graph-based systems

Design features and pipelines for large temporal and spatial datasets

Build scalable training pipelines, feature engineering flows, and inference services

Deploy low-latency model endpoints supporting real-time fleet decisions

Implement model retraining workflows, drift detection, and automated performance monitoring

Establish experiment tracking, model versioning, and performance tracking practices

Conduct deep error analysis and optimize models for reliability and stability

Define evaluation metrics aligned with operational constraints: SLA accuracy, idle km, throughput

Mentor MLE I engineers and guide best practices in modeling and pipelines

Work closely with backend teams to integrate ML systems into microservices

Contribute to architecture decisions, documentation, and cross-functional planning

25 years of experience building and deploying ML models in production

Strong experience with time-series forecasting, spatial modeling, or mobility datasets

Proficiency with PyTorch or TensorFlow and production-grade Python

Experience with distributed data systems, training pipelines, and ML orchestration

Strong understanding of model deployment, inference optimization, and monitoring

Familiarity with geospatial models, H3/hex grids, or GNN-based architectures

Ability to break down ambiguous problems and own end-to-end systems

Strong communication with engineering, product, and operations teams

Ability to mentor junior ML engineers

Experience with real-time inference or high-throughput ML systems

Background in operations research, applied math, or geospatial analytics

Work with logistics, mobility, or high-frequency forecasting problems

Open-source contributions to ML or data infrastructure