Forward Deployed AI Engineer – Agentic AI & Harness Engineering

TribalScale helps enterprises become AI-native. We're looking for a hands-on Forward Deployed AI Engineer who can turn complex business problems into reliable, production-ready AI systems. What you'll do • Build AI agents, multi-agent systems, and automated workflows. • Develop agent harnesses, loops, state graphs, tool integrations, and orchestration. • Build applications in Python and/or TypeScript using LLM APIs, RAG, memory, context management, and model routing. • Implement evals, observability, monitoring, and reliability controls. • Integrate with enterprise applications, databases, and AWS, Azure, or GCP. • Work directly with clients from requirements and prototypes through production deployment. • Use tools such as Claude Code and Codex to accelerate delivery. What you bring • Strong software engineering skills in Python and/or TypeScript. • Proven experience shipping AI agents and agentic workflows into production. • Practical knowledge of harness engineering, orchestration, tool calling, and evals. • Ability to solve ambiguous problems independently and communicate clearly with clients. Bonus: LangGraph, OpenAI Agents SDK, Claude, multi-agent architectures, AI security, and enterprise deployments. We want engineers who can build reliable systems around AI models, evaluate their performance, and ship working solutions. To apply, include: • Examples of AI agents or systems you personally built, with your specific contribution. • Your approach to harnesses, loops, graphs, evals, and production reliability. • Your GitHub or relevant portfolio. • Your hourly rate, weekly availability, and earliest start date. • A 3–5 minute Loom demo of an agent you built, covering architecture, tools, orchestration, and evaluation. Use a sanitized example with no confidential client data. We're looking for builders who can take AI from prototype to production.