AI Engineer Mid-Level

We are looking for a hands-on AI Engineer - Mid-Level who can build, integrate, and support production-grade AI capabilities within enterprise applications.

This is an engineering-focused role that combines Python development, AI API integration, model evaluation, data workflows, and production-quality implementation. The AI Engineer will collaborate closely with Product, Backend, QA, and Infrastructure teams to deliver reliable AI features with strong safety, quality, and performance controls.

The ideal candidate should be comfortable working with real-world business data, APIs, AI models, and production workflows while following disciplined practices for data privacy, validation, monitoring, and responsible AI implementation.

• Bachelor's degree in Computer Science, Artificial Intelligence, Machine Learning, Software Engineering, or a related field is preferred.
• 3–6 years of professional software engineering experience.
• Hands-on experience building and supporting AI/ML or GenAI-enabled applications.
• Strong Python development skills.
• Experience working with APIs, data workflows, and production systems.
• Strong analytical and problem-solving abilities.
• Ability to collaborate effectively with technical and cross-functional teams.
• Strong attention to detail and commitment to software quality, data privacy, and responsible AI practices.

1. AI Application Engineering

• Build and maintain AI-powered backend services and APIs.
• Integrate model inference services into existing product workflows.
• Develop structured prompts, schemas, and validation logic.
• Implement confidence scoring and decision-making logic for AI outputs.
• Develop fallback mechanisms for uncertain, incomplete, or failed AI responses.
• Implement human-review triggers for low-confidence or high-risk outputs.
• Ensure AI features are reliable, scalable, and suitable for production environments.

1. Model Evaluation and Optimization

• Evaluate AI models based on accuracy, latency, cost, reliability, and overall performance.
• Develop regression tests to identify changes in AI behavior resulting from prompt or model updates.
• Analyze model errors, hallucinations, edge cases, and unexpected outputs.
• Develop evaluation datasets and quality metrics for AI workflows.
• Recommend appropriate model-routing strategies based on task complexity and business requirements.
• Determine when to use rules-based processing, OCR/parsing, smaller models, or larger models.
• Continuously identify opportunities to improve AI performance, reliability, and cost efficiency.

1. Production Integration

• Collaborate with backend engineers to define and implement API contracts.
• Support workflow orchestration and asynchronous AI processing.
• Implement secure handling of AI model inputs and outputs.
• Develop appropriate validation and error-handling mechanisms.
• Support monitoring of AI services, including latency, failures, usage, and quality metrics.
• Troubleshoot production issues and support continuous improvement.
• Write clean, maintainable, well-tested Python code.
• Maintain clear technical documentation and operational handoff materials.

1. AI Safety, Guardrails & Responsible AI

• Implement output validation so AI-generated results cannot directly trigger critical business actions without appropriate checks.
• Design confidence thresholds and fallback paths for uncertain or unreliable results.
• Implement hallucination mitigation techniques where applicable.
• Apply prompt-injection mitigation patterns to protect AI workflows.
• Ensure sensitive information is appropriately redacted, tokenized, minimized, or protected before being processed by AI models.
• Track model, prompt, and dataset versions for evaluated AI workflows.
• Maintain audit logs covering model calls, decisions, failures, and human overrides.
• Implement tenant-aware processing and prevent sensitive data from being exposed through shared logs or monitoring systems.
• Support human-in-the-loop review for low-confidence, sensitive, or high-risk AI-generated decisions.
• Follow responsible AI and enterprise data-security practices throughout the development lifecycle.