Machine Learning Engineer – Agentic LLM & Workflow Automation

LLM Fine-Tuning & Prompt Engineering:

• Adapt state-of-the-art LLMs (e.g., GPT-style, LLaMA-based) to domain-specific tasks—data extraction from well logs, automated reporting, anomaly detection.
• Create, iterate, and optimize prompts and training pipelines for few-shot and reinforcement-based fine-tuning.

Agentic Workflow Design:

• Build multi-step “agents” that chain LLM calls, external APIs, and rule-based checks to fully automate routine subsurface tasks.
• Ensure agents handle error recovery, context management, and scalability under production loads.

Backend & Orchestration:

• Develop microservices in Python or C# to serve LLM agents via API, integrating with CMG’s data stores and simulation tools.
• Containerize and deploy these services, instrumenting them with monitoring, logging, and performance metrics.

Scalability & Governance:

• Collaborate with DevOps to scale inference horizontally, balancing cost, latency, and throughput.

Cross-Functional Delivery:

• Partner with UX designers to provide intuitive interfaces (chatbots, dashboards) for engineers to interact with agents.
• Lead Agile sprints, demos, and retrospectives—communicating progress and trade-offs to product and domain teams.

The above statements are intended only to describe the general nature of the job and should not be construed as an all-inclusive list of position responsibilities.