Chiri

Compensation

Full Description

Full Stack Software Engineer – Governed AI Product Development Why Chiri Chiri helps teams work smarter by embedding AI directly into how work actually happens—without sacrificing governance, security, or trust. Our platform, Chiri Brain, is a governed AI OS for humans and agents: a control plane that makes AI transparent, auditable, and enterprise-safe by design. Think of us as AI sherpas. We meet clients where they are, map the terrain across data, risk, and workflows, and help them reach the next ridge safely and sustainably—with full visibility into how AI decisions are made, executed, and governed. Role Overview We’re seeking a Full Stack Software Engineer to help build and evolve Chiri Brain, our flagship governed AI platform. This role is ideal for an engineer who enjoys taking features from ideation to production and is excited by AI systems that must operate under real-world enterprise constraints. You’ll work on: AI governance and control-plane capabilities Workflow orchestration for human + agent workflows Retrieval-Augmented Generation (RAG) systems Fully traceable, auditable data pipelines This is not a “black-box LLM” role. You’ll help design systems where every AI action is observable, permissioned, logged, and explainable—from prompt construction to tool execution and model selection. Expect a fast-moving environment with high ownership, close collaboration with product and clients, and a strong emphasis on building AI that enterprises can actually approve and deploy. Key Responsibilities Product Development (≈80%) Build Governed AI Systems: Design and implement core platform features that power Chiri Brain’s AI control plane, including execution tracing, persona enforcement, guardrails, and policy-aware inference. Agent & Workflow Orchestration: Develop orchestration logic for agent workflows—coordinating multi-step tasks, tool usage, and model calls while preserving full auditability and deterministic replay. RAG & Knowledge Systems: Build and evolve RAG pipelines that support shareable document collections, semantic search, metadata extraction, and cited answers—aligned with Chiri Brain’s “see everything” philosophy. Multi-Model AI Control: Implement abstractions that allow workflows to run across multiple models (e.g., OpenAI, Anthropic, local models), including parallel execution and comparison patterns. Backend & Frontend Development: Backend: Python services handling retrieval, orchestration, inference auditing, policy enforcement, and data isolation. Frontend: TypeScript interfaces that expose execution traces, permissions, workflows, and AI interactions in a clear, enterprise-friendly way. Feature Lifecycle Management (≈20%) End-to-End Ownership: Take features from concept through production—partnering with product, design, and clients to define requirements, scope sprints, and deliver incrementally. Governance-Driven Design: Translate governance requirements (security, compliance, auditability, data retention) into concrete product features rather than after-the-fact controls. Cross-Functional Collaboration: Work closely with stakeholders across HR, Finance, Security, and Engineering to ensure AI workflows align with real operational and regulatory needs. Agile Execution: Use tools like Linear or JIRA to manage work independently, communicate progress, and iterate quickly based on feedback. What You’ll Help Build AI control-plane primitives (personas, policies, permissions, execution traces) Workflow orchestration for humans + agents Fully auditable inference pipelines (no black boxes) Enterprise-grade governance features: RBAC/ABAC, PII detection, guardrails, and immutable logs RAG systems designed for trust, reuse, and collaboration, not one-off demos Required Qualifications Bachelor’s degree in Computer Science, Software Engineering, or a related field. 2–4 years of professional full stack engineering experience, with demonstrated ownership of production features. Strong proficiency in: Python (APIs, orchestration logic, data pipelines, AI frameworks like LangChain) TypeScript (React, Vite, or similar modern frontend frameworks) Hands-on experience with: RAG systems and vector databases (e.g., PGVector, Pinecone) LangChain or similar agent/orchestration frameworks Familiarity with enterprise AI requirements, including: Audit logging and execution traceability Privacy-by-design (PII handling, data isolation) Regulatory awareness (GDPR, HIPAA, SOC-style controls) Comfort using AI-assisted development tools (e.g., Cursor, Claude Code, Codex CLI) to accelerate delivery. Strong English communication skills for documentation, collaboration, and client-facing discussions. Ability to work independently while operating within structured agile workflows. Preferred Qualifications Experience building AI platforms, not just AI features. Exposure to agent systems, workflow engines, or orchestration frameworks. Familiarity with RBAC/ABAC, policy engines, or security-conscious backend design. Experience working with enterprise or regulated clients. What We Offer A fast-paced, fully remote environment with meaningful ownership. The opportunity to help define what “enterprise-ready AI” actually means. Access to cutting-edge AI tools, models, and internal playbooks. The chance to help clients achieve tangible outcomes—like 50% faster workflows—without compromising governance or trust. If you’re excited about building AI systems that enterprises can confidently say “yes” to, we’d love to hear from you.