Applied AI Engineer

Fuse Energy

About The Position

Fuse Energy is an energy startup focused on making energy abundant and affordable. It is building an integrated energy business spanning generation projects, hardware, grid infrastructure, real-time power trading, AI, and distributed energy installations in homes. The Applied AI Engineer will help build the Product Factory, an agentic pipeline that takes software from requirements and design through code review, deployment, and monitoring, with human review where risk warrants it. The role also ships customer-facing features using VLMs and LLMs and builds AI tools and agents to help teams across Fuse, including customer support, warehouses, and back-office functions. This is an applied engineering role, with ownership of work from specification through production. The engineer will collaborate with trading, operations, and backend teams so AI systems account for real-time market conditions, energy pricing, demand, and weather forecasts, and will monitor production systems to keep them reliable.

Requirements

At least 3 years of experience; requirements vary by role level.
Experience owning an applied AI system end to end, ideally in a startup or small team, and ability to explain what broke and how success was assessed.
Experience as a software engineer who moved into applied AI or as an ML engineer who writes production code.
Strong Python or similar programming skills.
Experience deploying LLMs or VLMs into production.
Solid understanding of cloud technologies, containerization, and scalable AI applications.
Analytical, rigorous, and quick to learn.

Nice To Haves

PyTorch or TensorFlow experience.
Forecasting, weather, and demand modeling experience.
Knowledge of energy markets and trading.
Fine-tuning or reinforcement learning experience.
Experience with vision, document, or NLP pipelines.
Experience with agent runtimes.

Responsibilities

Ship customer-facing AI features, such as personalized energy recommendations and onboarding based on an energy bill.
Build LLM applications and agents for support, engineering, warehouse, and back-office teams, including multi-agent systems and durable workflows such as Temporal.
Build supporting engineering, including cloud infrastructure, integrations, and access control that limits users and agents to appropriate resources.
Own evaluation through datasets, golden answers, human and LLM judges, and A/B tests to determine whether changes help.
Choose models based on performance, latency, and cost, and revisit choices as models improve.
Work with trading, operations, and backend teams so models reflect real-time market conditions, energy pricing, demand, and weather forecasts.
Monitor AI systems in production and maintain their reliability.
Own applied AI work end to end, from specification through production.