Posted:
1/8/2026, 7:15:23 AM
Location(s):
California, United States ⋅ Santa Ana, California, United States
Experience Level(s):
Senior
Field(s):
AI & Machine Learning
WHAT YOU'LL DO:
Design, build, fine-tune, and deploy state-of-the-art machine learning and large language models at scale, supporting millions of daily predictions with a strong focus on accuracy, latency, compute efficiency, and cost optimization.
Develop end-to-end ML and LLM pipelines, covering data ingestion, scripting, automated workflows for OCR, model training, evaluation, and post-processing in production environments.
Build and operationalize LLM fine-tuning pipelines, applying a range of model adaptation techniques including full fine-tuning, LoRA (Low-Rank Adaptation), prompt-based methods, and Direct Preference Optimization (DPO).
Design and experiment with novel LLM architectures, balancing model size, computational efficiency, memory constraints, and deployment requirements.
Optimize LLMs for production deployment through model quantization, compression, and teacher–student architectures, enabling efficient inference in resource-constrained environments.
Architect and deploy Retrieval-Augmented Generation (RAG) systems, leveraging vector databases, embedding services, semantic search, document chunking, indexing, and retrieval mechanisms using frameworks such as LangChain, LlamaIndex, and commercial RAG platforms within GCP and Databricks.
Innovate in ML operations and evaluation, including automated ground-truth generation, continuous post-evaluation pipelines, and iterative feedback loops to systematically improve model performance over time.
Design and implement CI/CD pipelines for machine learning systems, ensuring high availability, reliability, low latency, and rapid iteration from experimentation to production.
WHAT YOU’LL BRING
5+ years of experience in machine learning engineering, with a proven track record of deploying and operating ML and NLP/LLM systems in production at scale.
Strong hands-on experience building full-stack ML systems, from data ingestion and automation to training, evaluation, deployment, and monitoring.
Deep expertise in LLM fine-tuning and adaptation techniques, including full fine-tuning, LoRA, prompt-based optimization, and preference-based methods such as DPO.
Practical experience designing and optimizing LLM architectures, with an emphasis on compute efficiency, memory usage, and real-world deployment constraints.
Demonstrated proficiency in model inference optimization, including quantization, compression, and distillation techniques for high-throughput, cost-efficient production systems.
Solid understanding and hands-on experience with RAG architectures, vector stores, embeddings, semantic search, chunking strategies, and retrieval workflows integrated with large language models.
Experience using modern LLM orchestration and RAG frameworks such as LangChain, LlamaIndex, and managed AI platforms within cloud ecosystems like GCP and Databricks.
Strong background in ML evaluation and MLOps, including automated evaluation pipelines, CI/CD for ML, and continuous improvement of deployed models.
Proficiency in Python and ML/AI development frameworks, with the ability to work in fast-paced, experimental environments and production systems simultaneously.
Pay Range: $126,100 - $212,658 Annually
This hiring range is a reasonable estimate of the base pay range for this position at the time of posting. Pay is based on a number of factors which may include job-related knowledge, skills, experience, business requirements and geographic location.
** Note that the following statements only apply to candidates who will be working from an unincorporated area within Los Angeles County. **
First American will consider for employment all qualified applicants, including those with arrest or conviction records, in a manner consistent with the requirements of applicable state and local laws (e.g., the Los Angeles County Fair Chance Ordinance for Employers and the California Fair Chance Act).
First American intends to conduct a review of an applicant’s criminal history in connection with a conditional offer. First American reasonably believes that a criminal history may have a direct, adverse and negative relationship with the following material job duties for this position potentially resulting in the withdrawal of the conditional offer of employment: handling of confidential, proprietary or trade secret information belonging to First American or its customers, administrating or facilitating financial transactions, and the ability to meet customer-imposed criminal history requirements.
Website: https://www.firstam.com/
Headquarter Location: Santa Ana, California, United States
Employee Count: 10001+
Year Founded: 1889
IPO Status: Public
Industries: Financial Services ⋅ Insurance ⋅ Property Insurance ⋅ Real Estate ⋅ Real Estate Investment