- Company Name
- Primary Talent Partners
- Job Title
- AI Engineer – LLM Integration & Agent Development
- Job Description
-
**Job Title**
AI Engineer – LLM Integration & Agent Development
**Role Summary**
Design, build, and deploy Python‑based AI services that integrate large language models with enterprise data. Lead the construction of data pipelines, vector‑search infrastructure, RAG workflows, and multi‑agent orchestrations for real‑time applications in a contract environment.
**Expectations**
- 36‑month contract with potential extension.
- Deliver clean, modular code, comprehensive documentation, and maintain high reliability.
- Collaborate with data scientists, platform engineers, and stakeholders to translate business requirements into technical solutions.
**Key Responsibilities**
1. Develop and maintain Python APIs linking data sources, function calls, and LLM endpoints.
2. Build data pipelines that convert structured and unstructured data into embeddings, including SQL extraction, document parsing, and metadata tagging.
3. Integrate vector databases (Pinecone, FAISS, Weaviate, Chroma) to support retrieval‑augmented generation (RAG) workflows.
4. Implement multi‑agent frameworks (LangChain, LlamaIndex, CrewAI, OpenDevin) for orchestration and tool‑use.
5. Deploy models and RAG services on AWS or equivalent cloud platforms, using Docker containers and CI/CD pipelines.
6. Evaluate prompts, select models, and conduct lightweight fine‑tuning or adapter techniques such as LoRA/PEFT.
7. Monitor performance, troubleshoot issues, and iterate on system design to meet SLAs.
8. Produce unit tests, code reviews, and maintain documentation for reproducibility.
**Required Skills**
- 3+ years of professional Python development.
- Experience with open‑source LLMs (Claude or similar) and LLM integration.
- Knowledge of vector databases and RAG pipelines.
- Hands‑on with agent orchestration tools (LangChain, LlamaIndex, CrewAI, OpenDevin).
- API development and cloud deployment (AWS, Docker).
- Data engineering fundamentals (SQL, document parsing, metadata tagging).
- Basic model evaluation, prompt engineering, and fine‑tuning (LoRA, PEFT).
**Required Education & Certifications**
- Bachelor’s degree in Computer Science, Data Science, or related field (or equivalent professional experience).