AI Engineer

Stack

Responsibilities

• ▹Run and optimize the self-hosted inference serving layer on own GPU hardware (vLLM, SGLang, TensorRT-LLM)
• ▹Optimize aggressively: tensor parallelism, quantization (FP8, AWQ, GPTQ), KV-cache and prefix caching, continuous batching, speculative decoding
• ▹Serve multiple models and features off shared hardware: multi-LoRA, routing, request scheduling
• ▹Improve latency, throughput, and GPU utilization across AI workloads
• ▹Build visibility: instrument performance and usage across AI surfaces
• ▹Ship in-app agent layer: proactive nudges, smart suggestions, agents that summarize, draft, schedule, and act
• ▹Build the substrate underneath: tools, memory, orchestration, guardrails, evaluation harnesses

Requirements

• ▹Technical and hands-on with infrastructure, running real systems on real hardware
• ▹Full-stack builder mindset spanning infra and app layer
• ▹Performance-minded: engineers latency, throughput, and efficiency deliberately
• ▹Rapid-prototyping with modern AI tooling (Claude Code, agent SDKs)

Soft skills

About the company

In Tandem builds technology across four brands (OurFamilyWizard, Cozi, FamilyWall, Custody Navigator) to help families stay organized and communicate well.