Harrison Clarke ยท 2 days ago
Senior AI Engineer
United States
Full-time
Remote
Senior Level
6+ years expHarrison Clarke is a U.S-based applied AI company specializing in real-time video understanding systems. They are seeking a Senior AI Engineer to lead the applied research and productionization of their video search and retrieval stack, connecting natural-language queries to high-dimensional video representations.
ConsultingDevOpsHuman ResourcesInformation TechnologyStaffing Agency
Hiring Manager
Bradley Green
Responsibilities
Design and build natural-language-to-video retrieval systems using state-of-the-art architectures (e.g., V-JEPA, CLIP, SigLIP, Video-LLMs, ViViT, TimeSformer)
Develop temporal localization and video summarization capabilities with fine-grained moment-level embeddings
Stand up vector search infrastructure (FAISS, Milvus, pgvector, Pinecone) with optimized sharding, caching, and hybrid retrieval strategies
Optimize GPU inference and serving pipelines using ONNX Runtime, TensorRT, or ROCm for low-latency performance
Drive multi-GPU training and distributed serving (FSDP, ZeRO, DDP, NCCL/RCCL) with strong understanding of parallelization and quantization techniques
Collaborate with MLOps, backend, and product teams to deliver production-ready AI features at scale
Define and track key retrieval and relevance metrics (R@K, mAP, nDCG) and run live A/B evaluations
Mentor junior engineers, document design decisions, and drive innovation through rigorous experimentation
Qualification
Required
6-10+ years of experience in machine learning or applied AI, with 4+ years focused on video understanding, multimodal retrieval, or transformer-based models
Proficiency in PyTorch and deep learning frameworks; experience with video backbones, contrastive training, and representation learning
Strong understanding of vector databases, ANN search (HNSW, IVF), and embedding pipelines
Demonstrated ability to ship high-performance AI systems with GPU optimization, ONNX/TensorRT, or ROCm pipelines
Experience with distributed training, CI/CD for ML, and scalable data pipelines (MLflow, W&B, K8s, Docker)
Excellent communication skills and a collaborative, low-ego approach to problem solving
Preferred
Experience with temporal detection, video tracking, or re-ID
Exposure to Video-RAG or structured retrieval (metadata + knowledge graph)
Background in real-time or edge inference systems
Interest in privacy-preserving or regulated AI systems
Benefits
Competitive base salary
Bonus
Equity
Company
Harrison Clarke
Harrison Clarke is the Leading Staffing & Recruiting Firm in XOps & Cybersecurity.
11-50 employees
Funding
Current Stage
Early StageLeadership Team
Firas Sozan
Founder & CEO
Company data provided by crunchbase