AI Engineer

About DeepHow

DeepHow is a Physical AI platform for industrial manufacturing, pharmaceuticals, and utilities that helps organizations capture expert know-how, turn it into dynamic work instructions, and drive verified execution on the front line.

The platform spans knowledge capture and sharing, AI-powered verification through Smart Compare and photo/video validation, and time and motion intelligence through guided workflows, SOP adherence, and real-time execution visibility. DeepHow supports customers from knowledge capture to verified execution, with strategic account expansion often centered on verification, AI-guided workflows, and time and motion intelligence.

The Role

We’re looking for an AI Engineer to own our AI pipeline end-to-end. You’ll take over our production ML stack, harden it, and ship improvements fast. Day one, your focus is MLOps and production deployment making sure our models are fast, reliable, and cost-efficient at scale.

This is a build-and-ship role, not research. If you like turning prototypes into production systems that real users depend on, read on.

What You’ll Own

• Our AI pipeline — ingestion, processing, inference, monitoring
• Deployment and scaling of LLM, VLM, and speech models in production (GCP)
• Latency, cost, and reliability optimization across the stack
• RAG pipelines, prompting, and evaluation frameworks
• Infrastructure and tooling to accelerate experimentation and shipping

Education & Experience

• Bachelor’s or master’s degree in computer science, Engineering, or a related technical field (or equivalent practical experience)
• 3–7+ years shipping ML/AI in production
• Strong Python; fluent in PyTorch or TensorFlow
• Hands-on with LLMs — prompting, fine-tuning, RAG, evals
• Solid MLOps chops: CI/CD for models, monitoring, cost optimization
• Experience deploying on GCP or AWS (GCP preferred)
• Comfort with vector DBs, embeddings, and retrieval systems
• Startup-speed execution

Nice to Have

• Video, speech, or multimodal AI experience
• MLflow, Kubeflow, Airflow, or similar
• Manufacturing or frontline workforce context
• Background shipping AI features in a SaaS product

Why DeepHow

• Real AI problems with real users, not demos
• Direct impact on frontline workers and industrial operations
• Small team, high ownership, fast cycles
• Shape the AI roadmap from the ground up