Full Stack Developer (Python, Java, Docker, Kubernetes)

Ignite•Suitland, MD

20d

About The Position

As a developer for the US Census Bureau, you will design, develop, and maintain the highly complex routing and scheduling algorithms that power the Mojo field operations control system. You will architect modular, scalable optimization services capable of handling both massive batch schedules and real-time, on-demand routing. You will ensure maximum operational efficiency and data-driven accuracy for critical national survey workflows.

Requirements

5+ years’ experience designing and implementing routing, scheduling, or optimization systems
Strong algorithmic background including graph theory, clustering, and heuristic/approximation techniques
Proficiency in Python or Java for building optimization services and supporting tooling
Experience developing scalable, high-performance services integrated with enterprise control systems (e.g., Mojo)
Demonstrated ability to design test harnesses validating both solution correctness and optimization quality
Experience implementing performance tuning, parallel processing, and large dataset handling
Familiarity with REST APIs, data pipelines, and system integration patterns
Experience instrumenting systems with logging, metrics, and monitoring for operational visibility
Ability to document algorithms, constraints, and operational procedures and collaborate across engineering and operations teams
US CITIZENSHIP REQUIRED

Nice To Haves

Docker
Kubernetes

Responsibilities

Design and implement vehicle routing and scheduling algorithms using OR-Tools to generate optimal field assignments
Develop clustering algorithms to group workloads geographically and minimize travel time/cost
Implement constraint models (capacity, time windows, skills, priorities, survey rules) within the optimizer
Integrate the optimizer with the Mojo control system APIs for job intake, execution, and results publishing
Build batch and real-time optimization modes to support both scheduled and on-demand routing
Refactor optimizer components into modular, extensible strategy interfaces for new routing heuristics
Implement data preprocessing pipelines to normalize locations, distances, and travel matrices
Develop automated test harnesses validating solution correctness against known datasets and constraints
Implement “solution quality” scoring metrics (distance, balance, SLA adherence, cost) to evaluate goodness of routes
Create regression benchmarks comparing new algorithm performance vs. baseline outputs
Add parallel processing and scaling support to handle large route sets and high agent counts
Instrument services with logging and metrics to track runtime, solver performance, and solution quality
Package optimizer services for CI/CD deployment with reproducible builds and environment configs
Document routing logic, constraints, integration points, and operational runbooks
Provide production support, tuning, and continuous improvement of optimization heuristics and performance