Technical reviewer

Amir Iqbal

Senior AI Systems Architect and Lead Technical Reviewer

About Amir

Amir Iqbal is a senior AI systems architect at Esaholic and its lead technical reviewer. He leads the checks that decide whether an AI system is ready for production, which means ready for real users. No machine learning pipeline, agent workflow, or data system goes live without peer review, type checking, and testing designed to find weaknesses.

His specialty is serving AI models at scale, which means running them so many users can get answers quickly and reliably. He works toward very low response times, with targets under 250 milliseconds for real time inference, where inference simply means the model producing an answer. He tunes how models are served, how requests are grouped together, and how repeated work is reused, so answers arrive faster and cost less to produce.

Amir also verifies multi-agent systems. When several AI agents work together, he reviews their step by step behavior to stop endless loops, memory leaks, and runaway costs. He checks that agents only use the tools and data they are allowed to use, and that one conversation cannot leak into another. After launch, he sets up monitoring, known as observability, that tracks speed, accuracy, and drift, meaning gradual changes in data or model quality, so problems are caught early.

Expertise

  • Model serving: running AI models so users get fast, reliable answers
  • Latency: the time between a request and an answer
  • Multi-agent verification: checking that teams of AI agents behave safely and stop correctly
  • Observability: monitoring live systems for speed, errors, and quality changes
  • Code audits: careful reviews of code and architecture for flaws

Articles on Esaholic

Written by Amir Iqbal:

Reviewed by Amir Iqbal:

Back to the full team