Software Developer / Engineer - Philadelphia, PA (Locals Only)

Software Developer / Engineer
Location: Philadelphia, PA 
Work Schedule: Hybrid 3 days on site, 2 remote

Position Overview

We are seeking a Software Developer / Engineer to help design and implement an on-premises Large Language Model (LLM) platform with Retrieval-Augmented Generation (RAG) capabilities. This role will focus on deploying open-source AI models, integrating vector databases, and building secure, enterprise-grade AI solutions in a private environment.

This is an excellent opportunity for a developer with hands-on experience in modern AI technologies who enjoys building scalable, high-performance systems.

Responsibilities

• Deploy and optimize open-source large language models (LLMs) such as Meta Llama 3 and Mistral/Mixtral in on-premises or private environments.
• Develop Python-based applications for LLM inference, prompt engineering, and model integration.
• Optimize CPU-based model inference through quantization and performance tuning.
• Design and implement Retrieval-Augmented Generation (RAG) (RAG) pipelines.
• Configure and manage open-source vector databases such as Qdrant, Chroma, Milvus, or pgvector.
• Generate and manage embeddings while implementing metadata filtering strategies.
• Support enterprise security requirements, including air-gapped deployments, access controls, data privacy, and audit logging.
• Produce technical documentation, deployment guidance, and knowledge transfer materials for internal teams.
• Build a working prototype integrating an LLM, vector database, and RAG architecture.

Required Qualifications

• Professional experience deploying open-source LLMs (e.g., Meta Llama 3, Mistral/Mixtral) in on-premises or private environments.
• Strong Python development experience.
• Hands-on experience with LLM inference, prompt engineering, and AI application integration.
• Experience optimizing CPU-based inference through model quantization and performance tuning.
• Experience with vector databases such as Qdrant, Chroma, Milvus, or pgvector.
• Proven experience implementing Retrieval-Augmented Generation (RAG) solutions.
• Understanding of enterprise security, data privacy, air-gapped environments, access controls, and audit logging.

Preferred Qualifications

• Experience with LangChain or LlamaIndex.
• Familiarity with Docker and Kubernetes.
• Experience with inference frameworks such as vLLM, llama.cpp, or Hugging Face Transformers.
• Experience with Rust, Go, or C++.
• Previous experience working in enterprise or regulated environments.

Deliverables

• Reference architecture and deployment guidance.
• Working prototype integrating an LLM, vector database, and RAG solution.
• Technical documentation and knowledge transfer to internal teams.