Senior AI Developer

About Crebrl AI & Rithm
Enterprises don't need more AI experiments. They need AI that shows up on the P&L.

At Crebrl AI, we are building Rithm — an AI-native system of systems that links business promises and contractual obligations to the core transactional engines where execution actually happens: ERP, CRM, and finance. Our mission is to ensure every contractual commitment is kept, evidenced, and realised across millions of enterprise data points and transactions.

The Role
As a Senior AI Developer, you will design and ship core AI capabilities on Rithm — multi-agent workflows, retrieval pipelines, and structured reasoning engines that turn messy enterprise data into dependable execution. You will own features end-to-end: from prompt design and ontology mapping to production deployment, evaluation suites, and customer outcomes.

Role Details:
• Title: Senior AI Developer
• Location: Pune, India · Hybrid
• Experience Required: 8+ years (with 2+ years shipping production LLM features)
• Reports To: AI Platform Architect

Key Responsibilities

1. Build Agents and AI Features
   • Architect multi-agent workflows, tool use, and protocol integrations (MCP).
   • Build structured extraction and deep reasoning pipelines over enterprise datasets.
   • Lead prompt engineering, agent skill development, and context-window optimization.

2. Make AI Accurate and Trusted
   • Implement advanced retrieval architectures: RAG, Vectorless RAG, and knowledge graph retrieval.
   • Design rigorous evaluation suites, regression gates, and automated red-teaming pipelines.
   • Ensure every model output is explainable, deterministic, and backed by verifiable evidence.

3. Ship to Production
   • Build daily with AI coding agents to maintain high velocity without sacrificing code hygiene.
   • Continuously tune latency, inference quality, and token/GPU costs.
   • Integrate AI agent workflows into enterprise backends (ERP, CRM, finance) via APIs and MCP.

What You Bring
• 8+ years of software engineering experience, including 2+ years actively shipping production LLM features.
• Strong proficiency in Python or TypeScript, asynchronous architectures, REST APIs, and data pipelines.
• Comprehensive LLM toolkit: function calling, agentic frameworks, RAG / Vectorless RAG, and MCP (Model Context Protocol).
• Hands-on familiarity with knowledge graphs, ontologies, and structured data extraction.
• An evals-first engineering mindset: you quantify accuracy, cost, and latency metrics before claiming a win.
• Practical experience building with AI coding agents daily while understanding their limitations.

Our Values (T·R·U·S·T)
• Transparent — We work in the open.
• Respect — For people and promises.
• Uncompromising — On quality and security.
• Speed — Ship, learn, iterate.
• Teamwork — People and agents, one team