Senior Python Engineer

Dispatch Network is building a sophisticated internal platform that models the real-time behaviour of our last-mile logistics operations. It is a core piece of engineering infrastructure — a high-fidelity environment that lets our product and intelligence teams test and iterate on the systems that control how orders are assigned, priced, routed, and settled, without touching live operations.

You will be the founding engineer on this system. You will own the architecture, define the integration contracts that every downstream team depends on, and take the platform from zero to a fully instrumented, multi-service environment.

• A Python-based event-driven simulation engine with a clean hexagonal (ports-and-adapters) architecture cleanly separating physical logic, decision logic, and operational service integrations.
• Domain models and port abstractions that serve as the stable API contract for the intelligence team, the backend platform team, and internal service integrations — frozen early so all teams can build in parallel.
• Local adapter implementations for a suite of pluggable decision and operational service ports covering assignment, routing, pricing, payout, and configuration.
• An asynchronous event pipeline built on AWS SNS/SQS via LocalStack — designed so internal service consumers work identically in local and cloud environments with no code changes.
• Redis-backed state stores, a durable run log, and a Docker Compose environment for the full local stack.
• Remote HTTP/gRPC adapters integrating with internal services, with circuit breaker patterns and a shadow mode for safe rollout.

• 3+ years of Python development with strong command of typing, dataclasses, abstract base classes, and modern Python idioms (3.10+).
• Demonstrated experience designing and implementing event-driven or message-driven systems (SNS/SQS, Kafka, RabbitMQ, or similar).
• Proficiency with hexagonal or clean architecture — you can explain ports, adapters, and dependency inversion without looking it up.
• Experience with discrete-event simulation (SimPy strongly preferred) or equivalent deep experience with async/concurrent Python systems.
• Hands-on AWS experience: SNS, SQS, and working knowledge of LocalStack for local cloud emulation.
• Comfortable with Redis (hashes, sorted sets, pub/sub) as a live-state store.
• Strong testing discipline: pytest, property-based tests, and a habit of writing ports as testable contracts.

• Prior work in logistics, ride-hailing, or delivery platform engineering.
• Experience integrating ML model serving APIs (gRPC, REST) into production or simulation pipelines.
• Familiarity with H3 spatial indexing.
• Exposure to OSRM or similar routing engines.
• Experience as a founding or lead engineer on a greenfield Python system.

This is a systems engineering role, not a data science role. We are not looking for someone to train models — we are looking for someone who can build the environment models train in. The ideal candidate thinks carefully about interface design, writes code that other teams can safely depend on, and is comfortable owning an entire subsystem end-to-end.

You should be the kind of engineer who reads an architecture document and immediately sees the edge cases, who writes the port ABCs before a single line of implementation, and who pushes back when a design decision would make testing harder.

• Python 3.14+
• SimPy
• H3
• Flask / SocketIO
• AWS SNS / SQS
• LocalStack
• Redis
• Docker
• OSRM
• gRPC
• pytest / mypy / ruff
• SQLite