AI Systems & Deployment Engineer

About the AI Systems & Deployment Engineer Role (Full-Time, Engineering, AI/ML Infrastructure) Role Overview: We seek an AI Systems & Deployment Engineer to lead the full lifecycle of large language model deployments—from model selection and fine-tuning to production scaling and continuous optimization. You will bridge cutting-edge ML research with enterprise infrastructure, building reliable, low-latency AI services that power real-world products. Key Responsibilities:

• Deploy and scale open-source and proprietary LLMs using modern inference engines (e.g., vLLM, TGI, TensorRT-LLM, ONNX Runtime).
• Establish and maintain CI/CD pipelines for ML models in production.
• Fine-tune and optimize models for low-latency inference, including quantization, distillation, and throughput benchmarking.
• Design advanced Retrieval-Augmented Generation (RAG) pipelines and integrate enterprise vector databases (e.g., Pinecone, Weaviate, pgvector).
• Build multi-agent systems orchestrating tool use, memory, and inter-agent communication for autonomous workflows.
• Collaborate with Infrastructure Engineers to optimize GPU clusters, reduce costs, and meet uptime SLAs.
• Leverage NVIDIA’s software stack (CUDA, TensorRT, Triton Inference Server) for hardware efficiency.

Required Qualifications:

• 3+ years in AI/ML software engineering, with 1+ year focused on production LLM deployments, fine-tuning, and RAG architectures.
• Hands-on experience with inference frameworks and containerized ML workloads on Kubernetes.
• Strong Python proficiency and MLOps tooling (MLflow, Weights & Biases, DVC).
• Deep understanding of transformer architectures and fine-tuning techniques (LoRA, QLoRA, RLHF).
• Ability to benchmark models for latency, throughput, and quality.

Preferred Qualifications:

• Experience with multi-agent frameworks (LangGraph, CrewAI, AutoGen).
• Familiarity with NVIDIA Triton Inference Server and GPU profiling tools.
• Background in distributed systems or high-performance computing.
• Prior work with enterprise vector stores and hybrid search architectures.

What We Offer: Competitive salary and equity, access to cutting-edge GPU infrastructure, and a collaborative team valuing engineering excellence in a flexible remote-friendly environment.