AI Full Stack Developer

CACI•Dayton, OH

2d•Onsite

About The Position

The AI Full Stack Developer designs, builds, and deploys end-to-end web applications that leverage artificial intelligence and large language models in a secure Department of War environment. The role combines frontend user-interface development, backend APIs, database architecture, and integration of AI pipelines. The position is on-site at Wright-Patterson Air Force Base, Ohio, and requires U.S. citizenship and residence in the Dayton, Ohio area.

Requirements

U.S. citizenship.
Must be located in the Dayton, Ohio area.
Advanced proficiency in Python and API frameworks such as FastAPI or Flask.
Experience building user interfaces with modern frameworks such as React or Vue, or AI-focused UI libraries such as Streamlit.
Strong experience with relational databases such as PostgreSQL and vector databases such as pgvector, Milvus, or Chroma.
Hands-on experience designing and implementing Retrieval-Augmented Generation workflows.
Experience containerizing applications with Docker or Podman and using Git for version control.

Nice To Haves

Understanding of deploying containerized applications in restricted, disconnected, or IL5+ environments without active external cloud connections.
Experience using the Hugging Face library or hub to evaluate, download, and implement open-weight models such as Llama 3 or Mistral locally.
Familiarity with model runtime tools such as vLLM or llama.cpp and quantization techniques such as GGUF.
Familiarity with AI orchestration frameworks such as LangChain or LlamaIndex.
Ability to implement AI guardrails to prevent off-topic outputs and mitigate hallucinations.

Responsibilities

Design, build, and deploy end-to-end web applications using AI and large language models.
Develop backend application logic and APIs that connect user interfaces to AI models.
Build user interfaces for chat experiences and search dashboards.
Design and manage relational and vector databases for application data and AI-searchable embeddings.
Implement Retrieval-Augmented Generation workflows so AI can securely search and reference internal documents.
Containerize applications and use version control to package them for stable deployment.