Principal Engineer – AI Infra & Inference

Principal Engineer – AI Infra & Inference

We are partnered with a Stealth AI Infra startup (backed by a Tier 1 AI Lab and advised by 2 of the world’s most prominent ML thought-leaders), who are hiring a Principal SW Engineer (genuine progression to HoE / Chief Engineer).

The business already have enterprise customer traction & are backed by Perplexity and the VC who led investments in Hashicorp & Rancher. They are advised by the ex-GM & MD of Microsoft Research. They have proprietary tech & are building next-gen enterprise multimodal search & analytics platform, combining graph-based retrieval, multimodal indexing and cutting-edge LLMs to deliver lightning-fast, context-aware search. They fuse graph theory with multimodal learning, and are cutting compute, data & CO₂ while boosting accuracy on real-world problems.

The founding team blends deep technical and enterprise expertise – ex-FAANG engineers & scientists from Tier1 Labs, product operators, and PhDs from top global universities. Guided by advisors from Microsoft Research, Berkeley, and UPenn. Building AI that scales responsibly, from the physical world up and unlock the next frontier in AI with graph foundation models for unstructured data. Working at the frontier of enterprise AI Infra, building systems that power multimodal understanding for enterprise orgs.

They are looking for a Principal (Founding) Engineer with progression to Head of Engineering & beyond. You will work directly with the founders to design systems that scale up inference for cutting-edge models, build efficient training loops, and operationalize self-hosted LLMs (Llama-70B, Mistral-8x22B, etc.) using GPU autoscaling and model-serving frameworks. Technical ownership and leadership trajectory from day one.

Key Experience Required;

• 5 Years’ experience in ML / Software Engineering
• Hands-on experience serving large language (and non-language) models in production (GPU orchestration, inference optimization)
• Experience sharing an ML model across multiple GPUs
• A builder’s mindset: you relish ambiguous requirements, pick the right tool, and ship

Please apply ASAP if interested!