Tech Lead Architect

Tech Lead Architect (AI Infrastructure & Distributed Systems) Open to candidates based in Europe/UK, and US. We are building next‑generation AI infrastructure of ultra‑fast model inference, scalable LLM hosting, evals, model routing, observability, and developer‑friendly APIs. Role Overview As the Tech Lead Architect, you will be responsible for the foundational architecture of our AI platform. You will work hands‑on while also guiding long‑term technical direction and building key systems that power our platform. This is a founding‑level role with extremely high ownership. What You’ll Do Architect the overall AI serving platform (model execution engine, routing, safety, observability). Design multi‑node LLM inference pipelines optimized for throughput, latency, and cost. Implement architectural frameworks that support thousands of concurrent model requests. Establish core engineering principles and technical direction. Define GPU cluster layout, scheduling strategies, sharding, and resource isolation. Lead decisions around vLLM, TensorRT‑LLM, DeepSpeed‑Inference or custom kernels. Architect distributed compute layers, RPC frameworks, autoscaling, and fault tolerance. Lead decisions across data plane, control plane, orchestration, and microservices structure. Build high‑availability systems that serve AI workloads reliably at scale. Work with product teams to define how developers interact with the platform. Architect logging, token accounting, rate limiting, and streaming protocols. Make key architectural decisions that define the company’s long‑term technical roadmap. Mentor and guide engineers (backend, infra, ML, frontend). Interview, hire, and help scale the engineering team. Work directly with the founders on strategy, vision, and roadmap. What We’re Looking For Must‑Have 7+ years experience in software engineering, infrastructure, or systems architecture. Strong experience with distributed systems + microservices; GPU programming, CUDA, or model inference; cloud infrastructure (AWS/GCP/Azure); Kubernetes, Ray, or container orchestration; high‑scale backend systems and APIs. Proven ability to architect large systems end to end. Experience with high‑performance systems (latency, throughput, batching, caching). Strong instincts around reliability, scalability, and cost‑efficiency. Nice‑to‑Have Experience building or contributing to inference frameworks (vLLM, TensorRT‑LLM, TGI). Deep understanding of LLM internals, KV cache, quantization, tensor parallelism. Experience with data streaming, tracing, profiling, or log‑based architectures. Experience with ML training, fine‑tuning pipelines, or HF ecosystem. Startup/founding experience or appetite for zero‑to‑one environments. Background in cloud cost optimization or infra financial modeling. Architect an entire LLM serving platform that rivals Fireworks.ai throughput. Build distributed multi‑node inference with near‑linear scaling. Design a routing layer that chooses the best model based on latency/cost/accuracy. Create a unified logging + observability system for AI workloads. Architect a fine‑tuning and evals platform integrated with the serving layer. Build a blueprint for expanding globally across multi‑region data centers. Why Join Us Build foundational systems for a new AI infra company. Solve extremely hard technical problems with massive impact. Work directly with founders who understand the tech deeply. Define the engineering culture and architecture. Fast execution environment with ownership over entire systems. Competitive salary + founder‑level equity. Seniority level Mid‑Senior level Employment type Full‑time Job function Information Technology Industries Staffing and Recruiting Referrals increase your chances of interviewing at Arc.dev by 2x. #J-18808-Ljbffr