Senior​/Software Engineer, AI

Position: Senior / Staff Software Engineer, AI
About Kindo
Kindo is an agent automation platform for Dev Ops and Sec Ops teams. We help organizations automate high-friction operational work using autonomous agents that run reliably, securely, and  platform supports deployment on-prem, in hybrid environments, or in the cloud, with enterprise-grade security controls from day one.
We’re a small, highly technical team with strong customer traction and real enterprise revenue. Engineers have direct ownership over critical systems and shape how the platform evolves.

The Role
You will design, build, and operate core systems that enable autonomous agents to function reliably in production. This is software engineering with AI at the center, not ML research and not chatbot wrappers. You’ll build production-grade agentic workflows, retrieval and memory systems, multi-model execution, and tool-calling integrations that interact safely with enterprise systems.
This is also frontier work. Many of the patterns for agentic systems are still emerging. You’ll explore new approaches, prototype quickly, and turn what works into durable production systems. At the same time, strong distributed systems fundamentals still apply. These systems must be reliable, secure, observable, debuggable, and maintainable under real-world conditions.
You will help decide what to build, not just how to build it. We expect engineers to build an understanding of our users, question requirements, and propose better solutions rather than simply implementing a spec.
We’re hiring at both the Senior and Staff levels. Senior engineers take systems end‑to‑end with high ownership;
Staff engineers additionally shape architectural direction, define the platform’s durable defaults, and identify the highest‑leverage abstractions and guardrails. We’ll calibrate level, scope, and compensation to your experience and performance during the interview process.
What You’ll Build

Agent execution systems, including autonomous task loops, scheduling, triggers, and control planes
Retrieval and memory architectures, including context management, long‑term memory, and structured memory
Multi‑model routing and orchestration across providers, balancing quality, latency, cost, and failure modes
Tool‑calling and integration frameworks for safe interaction with external services and enterprise environments
Reliability, security, and operability foundations, including evaluation, observability, failure isolation, and recovery paths
The enterprise platform experience itself, including usable interfaces that make building, deploying, and managing agents accessible to teams who are not AI experts
Admin and governance surfaces that give organizations clear visibility and control over their AI usage, including who is running what, cost and policy controls, and audit trails

How You Build
AI is a first‑class tool in how we engineer. You use AI across design, prototyping, implementation, testing, debugging, and incident response, and you continuously refine workflows that increase leverage without sacrificing quality. You pair that velocity with discipline: guardrails, verification, and architectural boundaries that keep systems safe as autonomy increases.
What We’re Looking For
We care far more about what you’ve built than what’s on your resume.
You:

Have built and operated complex backend or distributed systems in production
Have built LLM‑powered or AI‑native systems beyond demos, with real users and real constraints
Have strong judgment around reliability, security, observability, and failure modes
Are comfortable operating in ambiguous frontier areas and validating ideas through rapid iteration
Use AI as a core part of your development workflow, not as an occasional convenience
Operate with high ownership and autonomy and take systems end‑to‑end
Relentlessly ask “why” to get to the core problem and target persona before building, instead of checking requirements off a list
Live in the product daily and fix what annoys you without waiting for permission
Instrument and measure your own work, iterating on real adoption and impact instead of shipping and forgetting

Technical requirements:

Type Script required, Python strongly preferred
Strong…