Software Engineer, Algorithms & Infrastructure

Base Power Company • Austin, TX

8d • Onsite

About The Position

As a Software Engineer on the Markets team, you’ll build and automate the decisions our battery fleet and financial commitments run on, from real-time battery dispatch to long-term market participation. You’ll take ambiguous, high-stakes market and control problems and turn them into reliable, automated infrastructure: the simulators, models, and trading-strategy systems behind how our batteries buy, sell, and dispatch power. This role has direct, measurable impact on Base’s top and bottom line. You’ll ship production decision infrastructure with backtest automation and regression-safe releases, deliver measurable improvements to our battery dispatch and control policies, and help bring ML-based methods like model-predictive control and policy imitation into production as we expand into new market modalities and geographies.

Requirements

• Strong ability to systematize requirements, define success for ambiguous problems with multiple objectives, and propose pragmatic solutions.
• A world-class ownership mindset — you own outcomes, not just tasks.
• Hands-on experience with AI/ML and HPC infrastructure, from a major cloud provider (AWS, GCP, Azure) or specialized ones (Modal, Beam).
• Strong interest in applied math and machine learning methods.
• Experience solving MLOps or similarly challenging observability/monitoring/versioning problems.

Nice To Haves

• experience with decision-making under uncertainty for stateful systems.

Responsibilities

• Design, build, and maintain the infrastructure, platforms, and algorithmic components we rely on for decisions about battery control and financial market participation — simulators, statistical models, algorithmic trading strategies.
• Work closely with utilities, fleet, and market teams to enable our participation in power markets under new modalities and across additional geographies, domestic and international.
• Automate the processes the Markets team relies on daily: backtest automation, release processes that prevent regressions in energy arbitrage policy behavior, and iteration speed improvements.
• Build the monitoring, alerting, and reporting that tells us how our automated decision processes are performing.
• Improve our existing battery physical operation policies.
• De-risk, implement, and maintain ML-based methods (supervised learning, model-predictive control, policy imitation for dynamic systems like DAgger) for automated decision making.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior
• Education Level: No Education Listed