Staff Machine Learning / Operations Research Engineer

Burq is revolutionizing commerce from port to doorstep with Burq Intelligence and is building Dispatch OS for ML-assisted dispatch decisions. The Staff Machine Learning / Operations Research Engineer will define the technical direction and build machine learning and optimization systems for pricing, quote selection, forecasting, routing, and delivery operations. This hands-on individual contributor role will own ML and optimization architecture, roadmap, evaluation, deployment, and cross-functional technical leadership.

Responsibilities:

• Technical direction: own the ML and optimization roadmap for Dispatch OS, deciding which problems get ML, which get solvers, which stay heuristic, and in what order
• Architecture: design the end-to-end architecture for model serving, evaluation, and optimization so it scales with order volume, new providers, and new customers
• Hardest problems: lead the most ambiguous, high-stakes modeling and optimization problems from framing through production, and guide technical work across Engineering and Data
• Raise the bar: set standards for experimentation, evaluation, and production ML, and mentor engineers through design reviews, pairing, and code review
• Strategic partnership: work with Product and leadership to shape product strategy and identify where ML and OR create competitive advantage
• Quote selection and pricing: design and ship models for quote selection, dynamic pricing, and reliability scoring that measurably improve margin, win rate, and conversion
• Forecasting: build demand and volume forecasting models, including modern deep learning approaches where they outperform classical methods, to - help operations and customers plan driver and fleet capacity
• Optimization: move dispatch decisions from heuristics to solver-based optimization for batching, route optimization, and vehicle/fleet recommendation, using OR-Tools, Gurobi, or similar where it earns its keep
• LLMs and agents: apply LLMs and AI agents to dispatch workflows, such as extracting provider rules and rate terms from documents, and automating quote follow-ups and exception handling
• Evaluation and trust: build rigorous, replayable evaluation frameworks that let customers verify a model's performance against their own historical decisions before they trust it to act
• Constraint translation: turn real operational constraints (driver availability, provider rules, cost ceilings) into model requirements, scoring logic, and optimization formulations
• MLOps: design and own automated pipelines for training, deployment, and monitoring so models stay accurate in production at scale
• AI-native work: use AI tools daily to speed up experimentation, automate eval pipelines, and debug faster

Requirements:

• 9+ years in applied ML or ML engineering, including multiple years operating at the senior or staff level, with models shipped to and maintained in production
• Track record of setting technical direction for ML or optimization systems, where architecture decisions you made shaped a product or platform over multiple years
• Demonstrated ability to lead complex technical initiatives across teams without direct authority
• Track record of ML or optimization systems with quantified, company-level business impact (e.g., tens of millions in revenue, or major utilization or margin gains)
• Deep experience with decision, ranking, and scoring problems where model outputs directly drive a business action
• Strong quantitative and algorithmic reasoning, including combinatorial problems, constraint satisfaction, and algorithm design
• Hands-on experience formulating and solving optimization problems (LP/MIP, constraint programming, or VRP-style routing)
• Experience with time-series forecasting in production
• Hands-on experience deploying LLM-based systems in production, such as fine-tuned models, extraction pipelines, or agents
• Experience owning end-to-end ML pipelines and MLOps (training, deployment, monitoring, retraining)
• Comfortable with messy, incomplete, constraint-heavy operational data, and able to build models that honor hard business constraints rather than treating them as soft penalties
• Experience building evaluation frameworks that non-technical stakeholders can understand and trust
• ideally in pricing, logistics, marketplaces, or operations
• Experience with delivery/dispatch software, TMS platforms, or routing systems
• Production experience with commercial or open-source solvers (OR-Tools, Gurobi, CPLEX)
• Pricing or revenue management experience in aviation, fleet, or transportation
• Published work, patents, or open-source contributions in ML, OR, or pricing