AI Operating System Engineer (with Zafin)

Position Type: 6-Month Contract

Location: Remote (US-Based)

Duration: 6 Months (Potential for Extension)

Hourly Rate: $85.00 – $100.00 / hour (DOE)

MUST HAVE: TECH SKILL(S): Zafin Exp.************PLEASE APPLY IF YOU HAVE ZAFIN EXPERIENCE Please note: NO 3rd party partnerships nor sponsorship available by the client

Position Overview: We are seeking an experienced AI OS Engineer for a high-impact, 6-month contract initiative. In this role, you will lead the architecture and integration of our next-generation AI Operating System (AI OS)—a core orchestration framework designed to seamlessly manage autonomous agents, multi-LLM routing, context memory systems, tool execution, and local-to-cloud compute pipelines. Because this is a 6-month deliverable-driven contract, you will focus on turning architectural blueprints into production-grade infrastructure, executing real-time evaluation frameworks, and optimizing latency and compute costs.

Key Responsibilities: Design, build, and deploy agentic workflows, dynamic task schedulers, and execution runtime environments powering internal AI applications. Implement robust retrieval systems, long-term state persistence, vector databases (e.g., pgvector, Qdrant, Pinecone), and hybrid-search mechanisms to optimize agent context windows. Architect multi-model routing layers (e.g., Anthropic, OpenAI, open-source foundation models) for cost-efficiency, fallback management, and low-latency inference. Develop secure sandbox environments for tool execution, code generation, API calls, and agent safety protocols. Build evaluation harnesses to track model drift, execution accuracy, hallucination rates, and latency bottlenecks. Containerize and deploy AI OS infrastructure on cloud environments (AWS / GCP / Azure) using CI/CD pipelines.

Required Qualifications: 5+ years of production software engineering experience, with 2+ years focused on building agentic frameworks, multi-agent orchestrations, or LLM infrastructure. Advanced proficiency in Python, Type Script/Node.js, and modern async execution models. Hands-on expertise with agent architectures and orchestration frameworks (e.g., Lang Graph, Auto Gen, CrewAI, Llama Index, or custom in-house runtimes).Proven track record working with vector databases, embedding systems, and hybrid RAG implementations. Direct experience with Docker, Kubernetes, vLLM / Triton inference engines, and cloud platforms (AWS Sagemaker, GCP Vertex AI, or Azure ML).Mastery of RESTful/gRPC APIs, message queues (Kafka, RabbitMQ, Redis), and microservice architectures.

Preferred Qualifications: Experience with local LLM serving, quantization methods (AWQ, GGUF), and self-hosted foundation models (Llama, Mistral).Deep understanding of sandboxed execution environments (e.g., Web Assembly, Docker-in-Docker, E2B) for safe AI agent tool execution. Prior contract experience operating in fast-paced, 6-month delivery cycles with clear milestone check-ins. Contract Milestones & Deliverables Finalize system architecture, set up local/cloud runtime execution environments, and deploy the core orchestration layer. Integrate multi-agent tool execution, long-term memory state persistence, and guardrail protocols. Conduct system-wide evaluation harness benchmarking, latency/cost optimization, and handoff documentation for internal engineering teams.