Samiran Singha

I’m currently a Business Analyst (LLM Quality & Evaluation) at Turing, applying a structured QA mindset to AI output validation. I evaluate AI-generated results against multi-dimensional quality rubrics and maintain high precision across thousands of records. I detect issues like hallucinations, grounding failures, and incorrect source usage, then document them with clear, step-by-step written rationales.

Previously, as a Test Engineer at L.G. Soft India, I designed and executed systematic test cases for Amazon Alexa on webOS. I reproduced and documented defects in JIRA with precise reproduction steps and root-cause analysis, then supported release confidence through regression, API, integration, and exploratory testing. I also used Chrome DevTools for browser-level debugging, inspected network requests and console errors, and delivered defect-trend and quality-outcome reports to senior stakeholders.

Before that, I worked as an Associate - Internal Audit at Slice, where I conducted root-cause investigations of data anomalies across 1,000+ monthly financial records. I monitored data trends and flagged unusual spikes or drops, collaborating daily across US and India time zones with disciplined, evidence-based written communication. I also led data validation during the NESFB Bank system migration to ensure field-level accuracy and zero data loss across integrated platforms.

Across roles, I’ve built a documentation-first, quality-driven operating style—mirroring customer-issue triage and release gate checks. I’m an experienced user of AI-assisted workflows (Claude, ChatGPT, and other tools) to accelerate evaluation throughput, summarise issues, and support structured analysis. I’m fully available for remote full-time work and focused on consistently meeting SLA targets through careful, traceable QA.