AI Platform Engineer

Required AI Platform Engineer
Position Overview:
We're assembling a small-scale team of innovators committed to a transformative mission: advancing generative AI from conceptual breakthrough to tangible product reality. As an AI Platfrom Engineer, you will be a critical architect of the technological infrastructure that brings our most ambitious GenAI concepts to life, transforming our digital intelligence solutions through cutting-edge AI innovation.
Build advanced AI platform that operates both as a cloud SaaS and as a fully self-contained on-prem / edge deployment, designed for privacy-sensitive and security-critical environments, at the intersection of backend development, AI integration, DevOps, and open-source systems engineering.
You will be part of the core team responsible for adapting, hardening, and operating our SaaS architecture in on-prem and single-node environments (on Prem servers, laptops). Working closely with architecture and product management and play a key role in making complex AI systems deployable, reliable, and operable outside the cloud.
Key Responsibilities
Platform & Application Engineering:
Adapt cloud-native AI services to on-prem and edge deployments (single node, no managed cloud services).
Build and maintain full-stack components: Backend APIs (Python / Node.js), Lightweight UIs or internal tools when needed, Ensure services are stateless, configurable, and portable across environments.
AI & Open-Source Integration:
Integrate and operate open-source LLMs for:
RAG pipelines
Agentic workflows
Tool calling and orchestration
Work with: Embedding models, Vector databases (local and embedded modes), Analytical engines (e.g., embedded SQL / columnar systems), Optimize inference for CPU / single-GPU environments (quantization, batching, caching).
DevOps & Runtime Engineering (Strong Focus):
Package services into portable Docker containers usable in:
On-prem servers (Kubernetes)
laptop / edge devices
Implement in-process scaling strategies (worker pools, task queues, batching).
Build simple, reliable deployment and startup flows (no heavy orchestration).
Manage configuration, secrets, logging, and observability in constrained environments.
Systems & Reliability:
Design for: Offline operation, Limited resources, Predictable performance
Implement graceful degradation between: SaaS mode, On-prem server mode, Single-node / laptop device mode, Debug complex interactions across AI models, storage, and runtime systems.