AI Developer

This is a fully remote position, open to applicants in United States.

📋 Description

• Train and fine-tune Large Language Models (LLMs) utilizing supervised fine-tuning (SFT)

• Collaborate with open-source models like LLaMA, Mistral, Qwen, and similar architectures

• Create LoRA/Q-LoRA pipelines for efficient fine-tuning processes

• Develop and enhance data preprocessing workflows, encompassing tokenization and long-context management

• Utilize and extend Hugging Face Transformers & Datasets for both training and inference tasks

• Analyze and process structured and semi-structured data formats, including XML/XSD files

• Implement document parsing solutions for Office formats using python-docx and OpenXML

• Design and execute end-to-end Retrieval-Augmented Generation (RAG) pipelines for document-grounded question answering and knowledge retrieval

• Construct and sustain vector stores and embedding pipelines with technologies such as FAISS, Chroma, Weaviate, or pgvector

• Optimize retrieval strategies, including hybrid search, re-ranking, and chunking for domain-specific datasets

• Develop and manage Model Control Plane (MCP) server integrations to enable LLM access to tools, APIs, and external data sources

• Design agentic workflows utilizing MCP to ensure controlled, auditable access to internal systems and context

• Deploy, operate, and maintain models in fully offline and air-gapped environments

• Conduct model optimization and quantization using GGUF, GPTQ, AWQ, and bitsandbytes

• Build and sustain inference systems employing vLLM, TGI, and Ollama

• Enhance GPU utilization through CUDA, cuDNN, and VRAM-aware batching techniques

• Manage local CI/CD pipelines for machine learning models without reliance on cloud services

• Oversee local model registries, versioning, and artifacts management

• Ensure RAG and MCP components function effectively in offline and restricted network environments

• Develop Python backend services for machine learning training and inference workflows

• Collaborate with relational and vector databases for RAG storage solutions

• Utilize Docker and Git for development and deployment workflows

• Employ Azure DevOps for CI/CD processes, including local runners when necessary

⛳️ Requirements

• Strong proficiency in Python for backend and machine learning development

• Expertise in PyTorch or TensorFlow, along with scikit-learn and pandas

• Solid understanding of Postgres or MySQL databases

• Experience with Docker and Git version control

• Practical experience in LLM training, fine-tuning, and optimization techniques

• Familiarity with Hugging Face Transformers & Datasets

• Knowledge of XML/XSD and tools for parsing Office documents

• Experience in deploying models using vLLM, TGI, or Ollama

• Understanding of model quantization methods like GGUF, GPTQ, or AWQ

• Experience in GPU optimization and leveraging the CUDA ecosystem

• Proven experience in creating solutions for offline, on-premises, and air-gapped environments

• Practical experience in designing and implementing RAG pipelines, including embedding models, vector stores, and retrieval optimization

• Experience in building or integrating MCP servers

• Familiarity with advanced RAG techniques such as HyDE or multi-hop retrieval

• Ability to discuss complex technical topics with both technical and non-technical stakeholders

🏝️ Benefits

• Competitive salary and performance-based bonuses

• Flexible working hours and remote work options

• Continuous learning and professional development opportunities

• Access to cutting-edge tools and technologies

• Collaborative and inclusive work culture