Finding the visual connection.
Reverse image search using image embeddings and similarity-based retrieval to find visually related images.
I’m Sarvesh Pathak. I work on computer vision and speech AI, with a foundation in test automation and a habit of building products.

From learning patterns in images and audio to building reliable tests and useful web experiences.
Showing all 5 projects
Reverse image search using image embeddings and similarity-based retrieval to find visually related images.
Developing speech emotion recognition with CNNs and mel-spectrograms, spanning audio preprocessing, feature extraction and model training.
Python-based self-healing locator frameworks for Robot Framework and Selenium. Uses DOM context and similarity matching to recover broken XPath locators, with healing logs to support debugging.
ELVOR PARFUMS is my e-commerce web project: responsive product browsing, shopping-cart and checkout flows, authentication and customer account experiences.
A market analytics dashboard for Indian equities, with live updates, charting, paper-trading features and AlphaBot, a local-LLM market assistant. Includes Playwright API, UI and end-to-end tests.
A progression from analytics to automation, and now applied machine learning.
Computer vision and speech AI, working on reverse image search and speech emotion recognition.
Web and mobile automation, API and load testing, and self-healing locator frameworks. Selenium, Robot Framework, Playwright, Appium, Postman and JMeter.
Hands-on experience with Tableau CRM / Salesforce Einstein Analytics and data analysis.
I enjoy turning technical ideas into something useful. My quality engineering background gives me a practical perspective on machine learning: understand the problem, inspect the data, test the workflow, and keep improving.
Alongside my professional work, I build independent web products to explore the full path from an idea to a working experience.
Python · PyTorch · CNNs · Computer vision · Image embeddings · Speech processing
Selenium · Robot Framework · Playwright · Appium · Postman · JMeter
React · Next.js · TypeScript · Node.js · Express · Prisma
MTech, Software Systems — BITS Pilani WILP (ongoing)
Bachelor’s, Information Technology — MGM’s Dr. G.Y. Pathrikar College (2020–2023)
Connect with me to discuss machine learning, automation or product ideas.