Senior AI/ML Engineer (LLM)
Budget / SalaryHourly project
TypeFreelance project
LocationRemote
Posted9 hours ago
Location: [Remote / Hybrid / On-site] · Type: Full-time · Level: Senior
About the role You'll own the AI reasoning core of ScheduleForce: the pipeline that turns extracted drawing and scope data into a structured, logic-driven construction schedule. You'll design the agentic workflows, retrieval systems, and prompting strategies that let an LLM reason about scope, sequencing, and dependencies.
Responsibilities
Design and build LLM-powered pipelines that convert engineering scope into schedule activities, durations, and logic
Architect agentic/tool-use workflows and retrieval (RAG) systems over technical documents
Develop evaluation frameworks to measure schedule accuracy and reduce hallucination
Collaborate with the construction SME to encode planning logic into the system
Optimize latency, cost, and reliability of model calls in production
Requirements
5+ years software engineering, 2+ years building LLM applications in production
Strong experience with agent frameworks, RAG, prompt engineering, and tool/function calling
Proficiency in Python and modern LLM tooling
Experience building evaluation and testing pipelines for non-deterministic AI systems
Ability to translate messy domain logic into reliable AI workflows
About the role You'll own the AI reasoning core of ScheduleForce: the pipeline that turns extracted drawing and scope data into a structured, logic-driven construction schedule. You'll design the agentic workflows, retrieval systems, and prompting strategies that let an LLM reason about scope, sequencing, and dependencies.
Responsibilities
Design and build LLM-powered pipelines that convert engineering scope into schedule activities, durations, and logic
Architect agentic/tool-use workflows and retrieval (RAG) systems over technical documents
Develop evaluation frameworks to measure schedule accuracy and reduce hallucination
Collaborate with the construction SME to encode planning logic into the system
Optimize latency, cost, and reliability of model calls in production
Requirements
5+ years software engineering, 2+ years building LLM applications in production
Strong experience with agent frameworks, RAG, prompt engineering, and tool/function calling
Proficiency in Python and modern LLM tooling
Experience building evaluation and testing pipelines for non-deterministic AI systems
Ability to translate messy domain logic into reliable AI workflows
Apply on Freelancer →
Project sourced from Freelancer.com. Applications happen directly on the original platform — we never collect your data.