About NoteSpark

NoteSpark is an AI-powered call analytics product built as a connector on the LeadSquared marketplace. It converts recorded sales calls into structured intelligence: transcripts,

diarization, sentiment, BANT signals, and action items, and enables conversational querying over call history using RAG. Because the product’s core output is AI-generated, quality here means more than “does the button work”; it means validating that the intelligence itself is accurate, consistent, and trustworthy.

What you’ll do

- Own test strategy and execution for NoteSpark across its modules: configuration playground, ingestion pipeline, lead-level insights, reports, and dashboards
- Build and maintain automation suites for API and UI testing, integrated into our CI/CD pipeline
- Write BDD-style test scenarios (Gherkin/Cucumber or similar) in collaboration with product and engineering, so requirements and tests stay in sync
- Design test approaches for AI-generated outputs: evaluating transcription accuracy, diarization quality, summary correctness, extraction precision (action items, BANT), and regression across model/prompt changes
- Test event-driven flows: webhook triggers, queue processing, and end-to-end call analysis pipelines
- Log clear, reproducible defects and work closely with engineers through fix and verification
- Use AI tools day to day to accelerate your own testing work: generating test data,



drafting scenarios, and analyzing results

Must have

- 3–4 years of QA experience with a strong foundation in test automation (Selenium, Playwright, Cypress, or similar; API automation with Postman/RestAssured or similar)
- Hands-on experience with BDD frameworks: writing and maintaining Gherkin scenarios with Cucumber, SpecFlow, or equivalent
- Positive working knowledge of AI tools and comfort using them as part of the testing workflow
- Solid understanding of QA fundamentals: test design, regression planning, defect lifecycle, and CI integration
- Strong analytical skills and attention to detail, especially for outputs where “correct” isn’t binary

Good to have

- Experience testing AI/LLM-based applications: evaluating non-deterministic outputs, building golden datasets, prompt regression testing, or model output scoring
- Exposure to audio/speech products or data-heavy SaaS platforms
- Basic scripting ability (Python, JavaScript, or C#) beyond test frameworks
- Familiarity with AWS or cloud-based test environments

Why join

You’ll be the quality owner on a small, senior team shipping an AI product to enterprise customers. This is a chance to define how AI-based applications should be tested (a skill set the industry is only beginning to figure out) rather than following an established playbook.

📌 Senior Quality Assurance Automation Engineer (Bengaluru)
🏢 LeadSquared
📍 Bengaluru