About the Role
:
The Applied AI Engineer will design, develop, build and deploy on-premise AI solutions, integrating large language models (LLMs), retrieval-augmented generation (RAG), and intelligent agents into enterprise systems.
This role combines machine learning engineering, generative AI development, and prompt design to deliver secure, scalable, and business-aligned AI applications.
Key Responsibilities
:
- Model Engineering
: - Fine-tune and optimize open-source LLMs using techniques such as LoRA/QLoRA.
- Quantize and deploy models efficiently using frameworks like vLLM, Triton, or TensorRT.
Evaluate model performance with domain-specific benchmarks and metrics.
RAG & AI Development
:
- Implement embeddings, vector databases, and retrieval-augmented generation pipelines.
- Develop AI agents and workflows using frameworks such as LangChain, LangGraph, or Semantic Kernel.
Connect AI systems securely with enterprise platforms (ERP, CRM, IoT, PLM, file systems).
Prompt & Workflow Design
:
- Engineer prompts, structured templates, and chaining logic to improve reliability.
- Collaborate with business teams to tailor AI outputs to specific use cases and compliance requirements.
Implement guardrails, grounding, and refusal rules to reduce hallucinations and ensure safe outputs.
Collaboration & Delivery
:
- Work with Data Engineers and MLOps teams to integrate clean data pipelines.
- Partner with domain SMEs to align AI solutions with business processes.
- Document workflows, best practices, and provide internal training on AI usage.
Required Skills
:
- Strong proficiency in Python and ML/AI libraries (PyTorch, HuggingFace, Transformers).
- Experience with LLM fine-tuning, optimization, and inference serving (LoRA, vLLM, Triton).
- Hands-on expertise with vector databases (Qdrant, Milvus, pgvector) and RAG pipelines.
- Knowledge of agent frameworks (LangChain, LangGraph, Semantic Kernel) and tool integration.
- Practical experience in prompt engineering and structured output design.
- Familiarity with on-prem AI deployment, containerization (Docker/Kubernetes), and API integrations.
- Strong problem-solving mindset and ability to translate business needs into AI solutions.