Kerja Sepenuh Masa, Software Engineer (Smart Scheduler) di BIPO - Maukerja

About the Role 

BIPO's Smart Scheduler team is making waves in the AI Rostering space. We are looking for a strong engineer (First Class Honours or Second Upper from top universities preferred) to join our growing team. This role gives you exposure across demand forecasting, optimisation, machine learning and software engineering work. You will spend time building context in different areas before focusing more deeply on where you can create the most impact. The path is intentionally flexible and will be shaped together based on business needs, your strengths, and where the strongest fit emerges. 

The team works across operations research, applied machine learning, and production engineering. This is a technical role that requires depth in both maths and code, not just building models but turning them into reliable systems that can run in production. 

What You Will Work On 

Demand Forecasting 

• Build forecasting models (LSTMs, transformer-based architectures) to estimate staffing needs. 
• Develop and maintain the feature pipeline, incorporating signals such as workload trends, historical demand, planned absences, calendar effects, business events, and other operational drivers. 
• Design evaluation frameworks that reflect business impact, not just model accuracy. 
• Translate forecasting outputs into reliable inputs for the optimisation engine, with clear assumptions, confidence levels, and fallback behaviour. 
• Work with product, operations, and engineering teams to continuously refine forecasting logic as business rules, service models, and workforce constraints evolve. 

Roster Optimisation 

• Formulate and implement constraint satisfaction and combinatorial optimisation problems using Google OR-Tools (CP-SAT solver and routing libraries). 
• Translate governance rules, collective agreements, and soft preferences into well-posed constraint models. 
• Design scalable constraint models, including symmetry breaking, constraint relaxation, branching strategies, and objective weighting, to achieve strong solution quality within practical runtime limits. 
• Develop reinforcement learning approaches to support the optimisation engine, such as action selection, repair recommendations, demand adjustment, or search guidance. 

Platform & Infrastructure 

• Build production-ready machine learning model and optimisation pipelines that are reliable, observable, and maintainable. 
• Contribute to CI/CD practices for model releases: versioning, rollback etc. 
• Work with data engineers to keep feature stores current and handle upstream schema drift gracefully. 

Product Collaboration 

• Work directly with business and operations stakeholders to understand requirements, clarify trade-offs, and translate real-world problems into product and technical solutions. 
• Participate in discovery sessions to pressure-test new feature ideas against technical and data feasibility before they enter the roadmap. 

Who We Are Looking For 

Background 

• First Class Honours (or Second Upper from top universities) preferred: Degree in Computer Science, Data Science, Statistics, Operations Research, Applied Mathematics, or a closely related quantitative discipline with a substantial coding component. 
• Equivalent demonstrated experience is equally welcome, especially if you have built technical systems involving forecasting, optimisation, or decision automation. 

Technical Skills 

• Strong Python engineering fundamentals, including clean code, testing, debugging, code review, and performance-aware implementation. 
• Hands-on experience training and deploying deep learning models in PyTorch, including recurrent and attention-based architectures (LSTM, GRU, Transformer variants). 
• Working knowledge of Google OR-Tools or a comparable constraint/mathematical programming framework (CPLEX, Gurobi, CP-SAT). 
• Familiarity with cloud platforms such as AWS, GCP, or Aliyun. 
• Comfortable working with containerised workloads using Docker, Kubernetes, or equivalent deployment environments. 

Nice to Have 

• Comfortable working across the ML lifecycle, including data preparation, feature engineering, experiment tracking, model packaging, evaluation, and serving. 
• Experience with workforce management, rostering, scheduling, logistics, supply planning, or other operational decision systems. 
• Exposure to reinforcement learning or learning-to-optimise frameworks. 
• Solid grounding in probability, statistics, linear algebra, and optimisation. 
• Ability to work directly with business or operations stakeholders to clarify requirements and translate ambiguous problems into technical solutions. 
• Experience using AI coding tools or agentic development workflows to accelerate implementation, debugging, refactoring, testing, or technical documentation.