What you will do:

Architect and build

• Design and implement core platform services that power Leena AI’s agentic workflows—agent orchestration, context management, tool routing, and evaluation infrastructure.
• Own infrastructure decisions across compute, deployment, and reliability for production AI workloads.

Operate as an AI-native engineer

• Use AI coding agents (Claude Code) as your default working mode—not occasionally, but as the foundation of how you build.
• Run multiple AI agent sessions in parallel. Write structured specifications and context files (Claude.md, rule files) that enable AI to produce high-quality first drafts.
• Verify, debug, and harden AI-generated code. You are the quality gate—not a passenger.

Lead cross-cutting initiatives

• Work on strategic projects that cut across product, platform, and infrastructure.
• Drive technical standards for AI-native development across the engineering org. You’ll be a force multiplier, not just a contributor.
• Mentor other engineers in AI-native workflows and help accelerate the team’s transition from AI-assisted to AI-native.

What we’re looking for:

We evaluate on following dimensions. Raw coding speed is not one of them.

Systems and architectural judgment

• 5+ years of professional software engineering experience building production systems at scale.

• Fluency in at least one modern stack (TypeScript/Node,

  Python,

  Go).

• You can decompose an ambiguous problem into well-bounded components, define interfaces, and make durable architecture decisions under uncertainty.

• You understand distributed systems, API design, data modeling, and production reliability—not from textbooks, but from having been paged at 2 AM.

AI tool proficiency (agent leverage)

• Demonstrated daily use of AI coding tools (Cursor, Claude Code, or equivalents) in production workflows—not toy projects.
• You structure problems for AI consumption: writing specs, context files, and decomposed tasks that let AI agents produce high-quality output.
• You know when AI helps and when it doesn’t. You can articulate failure modes, model selection trade-offs, and verification strategies.
• Experience with context engineering—managing prompt context, rule files, codebase documentation—to maximize agent effectiveness.

Verification and quality instinct

• Strong testing discipline—you treat TDD and automated testing as an AI supervision mechanism, not just a quality practice.
• You review AI-generated code with the same rigor (or more) that you’d review a junior engineer’s PR.
• You catch subtle issues: security vulnerabilities, N+1 queries, race conditions, poor error handling—especially in AI-generated output.

Communication and collaboration

• Clear written and verbal communication. You can explain architectural decisions to both engineers and business stakeholders.
• You write specifications that are precise enough for both humans and AI agents to execute against.
• Comfortable working directly with senior leadership and influencing technical direction.