Note: The job is a remote job and is open to candidates in USA. Orbien LLC is seeking a Principal Machine Learning Engineer to lead the development of advanced agentic AI and large language model systems. The role focuses on architecting multi-agent systems, optimizing retrieval-augmented generation pipelines, fine-tuning and evaluating models, building scalable ML services, and establishing standards for evaluation, observability, safety, and compliance.
Responsibilities
- Architect and implement multi agent systems capable of planning, tool use, and coordinated task execution
- Design and optimize RAG pipelines including embeddings, hybrid retrieval, reranking, and context window strategies
- Fine tune and evaluate small, medium, and large language models for domain specific reasoning and summarization
- Develop prompt engineering frameworks, guardrails, and automated evaluation suites for agent reliability
- Build scalable ML services and APIs for production deployment in distributed environments
- Collaborate with product, engineering, and domain experts to translate complex workflows into agentic AI solutions
- Establish best practices for model evaluation, observability, safety, and compliance
- Mentor DS/ML engineers and contribute to long term AI strategy and architecture
Skills
- 6–12+ years in Data Science / ML Engineering, with deep experience in LLM based systems
- Proven experience building agentic architectures (planner executor, tool use agents, ReAct style reasoning)
- Strong background in RAG, embeddings, retrieval optimization, and evaluation
- Expertise in NLP, transformers, deep learning, and model fine tuning
- Proficiency with PyTorch, HuggingFace, LangChain/LlamaIndex, Ray, Kubernetes, and vector databases
- Experience designing production grade ML systems with monitoring, evaluation, and observability
- Strong communication skills and ability to lead technical direction
- Experience in enterprise search, knowledge management, or high compliance domains
- Experience with model distillation, LoRA/QLoRA, PEFT, and model compression
- Experience building evaluation frameworks for hallucination, grounding, and agent reliability
- Familiarity with knowledge graphs, symbolic reasoning, or hybrid neuro symbolic systems
- Publications, patents, or open source contributions in LLMs or agent systems
- Strong coding skills in Python 7+ years
- Be a natural problem solver, able to take a lead in collaborating to resolve issues
- Proficiency in IDE debugging : VSCODE and PYCHARM
- Have communication skills
- 5+ years of experience in AI and machine learning
- Deep understanding of machine learning algorithms, classification models, diagnostic testing of models
- Experience working directly and Transformer based architectures including BERT, RoBERTa, T5 etc. Nd familiarity with large language models and fine tuning
- Experience with conversational search / semantic search, reinforcement learning, prompt engineering, hallucination mitigation
- Working understanding of the business risks associated with applying LLM (LangChain) in a business
- Experience working with AWS, RAG, SageMaker, SQL
Benefits
- Remote work
Company Overview
Company H1B Sponsorship