Software Engineer

Job Summary

The Software Engineer designs, builds, and maintains robust reinforcement learning sandboxes and reference systems using various languages (Java, Node.js, Python, Go, Rust) to train next-generation AI. Core daily responsibilities include coding isolated backend testing environments, writing strict unit and integration tests, performing security-focused code reviews, solving complex real-world programming tasks, and documenting technical reasoning to refine model intelligence.

Foundational Capabilities & Prerequisites

• Polyglot Backend Expertise: Strong experience in backend software development leveraging Java, Node.js, Go, Rust, or Python.
• Scale Architecture: Verifiable track record building scalable backend systems, microservices, and robust APIs.
• Testing Rigor: Masterful command of automated testing frameworks. Unit and integration test coverage is central to this engineering role, never an afterthought.
• Diagnostic Proficiency: Deep experience handling code debugging, runtime performance optimization, and memory profiling.
• Distributed Collaboration: Ability to work effectively and asynchronously within a distributed team environment, writing down engineering decisions with total clarity.
• Communication Fluency: Excellent written and verbal English communication mechanics.

Preferred Cybersecurity & Technical Multipliers

• Cybersecurity / SecOps Exposure: Highly preferred background in secure software development, application security (AppSec), penetration testing, or vulnerability assessment.
• Security Standards Mastery: Working familiarity with OWASP guidelines, CWE (Common Weakness Enumeration) registries, or security-focused code audits.
• IDE Ecosystem Familiarity: Comfortable working across multi-faceted toolsets, such as Eclipse, terminal environments, and containerized runtime systems.
• Advanced Data Frameworks: Familiarity with data-driven, large-scale technical projects, or baseline artificial intelligence mechanics is considered a plus.

Technical Workflow Architecture

As an AI Evaluation Software Engineer at micro1, your engineering routine balances production-level coding with adversarial environment design:

• Sandbox Orchestration: Compiling an isolated, fully reproducible container environment that simulates a messy legacy enterprise application, complete with structural dependencies and logic bugs.
• Golden Reference Engineering: Writing the perfect, production-grade target solution (the "golden reference") that fixes the environment's flaws while meeting strict performance, scalability, and security parameters.
• Secure Code Auditing: Embedding subtle security vulnerabilities (such as SQL injection, broken authentication, or path traversal) to see if the AI model can accurately detect the flaw, describe the risk vector using OWASP/CWE terminologies, and patch it securely.
• Deterministic Test-Driven Evaluation: Writing aggressive automated test suites that execute against the AI model's code outputs to ensure it doesn't break peripheral system architecture or downstream microservices.

Key Responsibilities

1. Reinforcement Learning Environment Synthesis

• Build and maintain isolated reinforcement learning environments that exercise real-world backend programming tasks, feature creation, and refactoring simulations.
• Write robust backend services and the supporting infrastructure code that powers those sandboxes using Java, Python, Node.js, Go, or Rust.
• Generate reproducible problem sets and create the definitive golden reference solutions used to benchmark model capabilities.

2. Rigorous Test Automation & Quality Assurance

• Write exhaustive unit and integration tests to wrap around every simulated environment, ensuring test coverage serves as a core validation pillar.
• Review code payloads for absolute correctness, runtime efficiency, long-term maintainability, and structural security compliance.
• Provide clear, granular, and technically sound review feedback to peer engineers within the distributed workspace.

3. Asynchronous Problem Solving & Security Mapping

• Solve diverse backend programming tasks independently and document your step-by-step technical reasoning so it can be ingested to scale up AI model quality.
• Infuse secure coding principles, vulnerability remediation steps, and security-focused code reviews directly into the environment design.
• Coordinate asynchronously within a globally distributed engineering network, logging code modifications and architectural decisions clearly in writing.

Core Competencies & Success Indicators

• Architectural Containment: Crafting sandbox parameters so tight that model code executions can be measured deterministically without side effects.
• Adversarial Design Thinking: Simulating realistic, complex developer scenarios—including performance regressions and legacy debt—to comprehensively stress-test model reasoning.
• SecOps Code Validation: Evaluating if an AI tool's proposed solution violates corporate data governance, access controls, or secure development lifecycles.

Skills Required:

• Computer / Software / It / Data