AI Full Stack Developer

CACI International•Dayton, OH

4d•$63,300 - $129,700•Onsite

About The Position

The candidate will be responsible for designing, building, and deploying end-to-end web applications that leverage Artificial Intelligence (AI) and Large Language Models (LLMs) within a secure Department of War (DoW) environment. This requires standard full-stack development skills (frontend UI, backend APIs, and databases) combined with the ability to integrate modern AI pipelines.

Requirements

Advanced proficiency in Python and API frameworks (e.g., FastAPI, Flask)
Experience building clean, intuitive user interfaces using modern frameworks (e.g., React, Vue) or AI-focused UI libraries (e.g., Streamlit)
Strong experience with relational databases (e.g., PostgreSQL) and Vector Databases (e.g., pgvector, Milvus, Chroma)
Hands-on experience designing and implementing Retrieval-Augmented Generation (RAG) workflows
Experience containerizing applications using Docker/Podman
Experience utilizing version control (Git)

Nice To Haves

Understanding of how to deploy containerized applications in restricted, disconnected, or IL5+ environments without relying on active external cloud connections.
Experience utilizing the Hugging Face library/hub to evaluate, download, and implement open-weight models (e.g., Llama 3, Mistral) locally.
Familiarity with model runtime tools (e.g., vLLM, llama.cpp) and quantization techniques (e.g., GGUF) to run large AI models efficiently on standard DoD hardware.
Familiarity with frameworks like LangChain or LlamaIndex to accelerate the development of complex AI agent workflows.
Ability to implement system "guardrails" to prevent off-topic outputs and mitigate AI hallucinations, ensuring secure and accurate mission-support tools.

Responsibilities

Designing, building, and deploying end-to-end web applications that leverage Artificial Intelligence (AI) and Large Language Models (LLMs) within a secure Department of War (DoW) environment.
Integrating modern AI pipelines.
Building the core logic and connect user interfaces to AI models.
Building clean, intuitive user interfaces for chat interfaces and search dashboards.
Managing both application data and AI-searchable embeddings.
Designing and implementing Retrieval-Augmented Generation (RAG) workflows to allow the AI to securely reference and search internal documents.
Containerizing applications using Docker/Podman and utilizing version control (Git) to package applications for stable deployment.