Dhruv Sharma
About
Detail
AI Engineer | GenAI & LLM Systems | AI Agents | RAG | Production-Grade Automation | Ex-SDET
India
Résumé
Jobs
verified_user
0% verified
Co-Founder & AI Systems EngineerKinoJan 2026 - Current (8 months)Building KINO, an autonomous AI engineering platform designed to help engineering teams plan, build, review, repair, and manage software through coordinated AI agents. Leading the architecture and development of: - multi-agent orchestration - engineering planning systems - autonomous code review - quality gates - self-healing repair loops - knowledge graph and project memory - GitHub-native engineering workflows - enterprise governance and billing - production execution infrastructure The platform is designed as an AI operating system for autonomous software engineering.- USoftware Engineer in TestUolo EdTechNov 2025 - Aug 2026 (10 months)Engineered Python/Selenium suite covering 200+ workflows, cutting regression cycle time by 60% and eliminating manual execution on all critical user paths. Eliminated 40% of flaky tests via root-cause analysis, improving CI/CD signal reliability and unblocking 3 engineering squads; validated 50+ REST API contracts pre-release.
- IAI Engineer (Independent)Independent Applied AI EngineerOct 2025 - Jan 2026 (4 months)Focused on building reproducible, production-aware ML systems rather than academic prototypes. • Built defect risk prediction models using real-world software metrics • Developed ML-driven test case prioritization system (APFD evaluation) • Applied class-imbalance learning techniques for real-world datasets • Published reproducible applied AI research with public datasets Emphasis: Applied ML + validation + engineering rigor.
- Software Test Engineer (AI & Automation)Clikon technologies Pvt LtdAug 2023 - Oct 2025 (2 years 3 months)Worked at the intersection of automation engineering and applied AI, enabling reliable production releases for large-scale SaaS platforms. Key Contributions: • Designed scalable automation frameworks using Playwright & Selenium for high-traffic web applications • Integrated AI-assisted testing into CI/CD pipelines (Jenkins) • Built Python-based backend validation systems for API and system-level quality assurance • Developed intelligent regression workflows for defect trend analysis and release risk prediction Impact: • Reduced regression failures • Accelerated CI/CD feedback cycles • Improved production release confidence
Education
verified_user
0% verified
- Bachelor of Technology, Computational ScienceDr APJ Abdul Kalam Technical UniversityAug 2020 - Jul 2024 (4 years)
Projects (professional or personal)
verified_user
0% verified
- EE-Retail Inventory & Order Management PlatformJan 2024 - Current (2 years 8 months)Domain: E-Commerce / Retail Technology Role: QA Automation Engineer Tech Stack: Selenium (Java), TestNG, POM, Postman, Jira, Jenkins, MySQL Project Description: The E-Retail platform is a large-scale inventory and order management system used by warehouses to track product availability, packing, shipping, and order status in real time. The system manages saleable stock, packed inventory, shipment flow, and order lifecycle, ensuring accurate fulfillment across multiple warehouses. Responsibilities & Impact: • Designed and implemented end-to-end UI automation using Selenium WebDriver with Java and TestNG (POM framework) to validate critical business workflows such as order placement, inventory updates, packing, and shipment tracking. • Deve
- YYoutube-CloneApr 2024This YouTube Clone project is a web application built using React that replicates the core functionalities of YouTube. It provides users with a familiar interface and experience for watching videos, searching for content, and managing playlists. Features: Responsive Design,Video Playback,Search Functionality.
- PPassword GeneratorFeb 2024 - Mar 2024 (2 months)The combination of the symbols, alphabet and numbers make a strong password which is not easy guess my any one.
- WWeather AppFeb 2024This user-friendly weather app provides current and forecasted weather information for any location worldwide. Leverage the power of the Frontend OpenWeather API to retrieve real-time weather data, including temperature, humidity, precipitation, wind speed, and more
- EEPRS Distributor & Recharge Management SystemSep 2023 - Dec 2024 (1 year 4 months)Domain: SaaS / FinTech-style Platform Role: QA Automation Engineer Tech Stack: Selenium (Java), TestNG, Postman, Jira Project Description: EPRS is a Distributor and Recharge Management platform used by dealers to manage complaints, recharge transactions, and account corrections. The system allows incorrect recharges to be transferred to the correct account, making accuracy and data integrity critical. Responsibilities & Impact: • Automated UI and functional workflows using Selenium (Java + TestNG) to validate recharge, complaint handling, and account transfer scenarios. • Designed test strategies, regression suites, and data-driven tests to ensure transaction accuracy and system stability. • Performed API and backend validation to verify
- CCalculatorJun 2023 - Oct 2023 (5 months)it's a calculator for solving basic arthimetic operation and trigonometry function value. this is made by tkinter library.
Publications
verified_user
0% verified
- ZAI-Based Test Case Prioritization Using Defect Risk PredictionZenodo Open Research Repository FebThis research proposes an AI-based test case prioritization approach that integrates machine learning–driven defect risk prediction with automated testing workflows. Defect probabilities are first estimated at the software module level using static code metrics and a class-balanced Logistic Regression model. These risk scores are then propagated to test cases based on coverage relationships, enabling risk-aware test execution. The proposed prioritization strategy is evaluated using the APFD metric and demonstrates improved fault detection effectiveness compared to baseline execution orders such as random and original sequencing. Dataset: NASA CM1 (PROMISE repository) Focus areas: – Applied machine learning for defect risk predictio
- Zsoftware-defect-prediction-ml-test-automationZenodo Open Research Repository JanThis project explores the application of machine learning techniques for predicting defect-prone software modules using static code metrics. A Logistic Regression model with imbalance-aware learning is evaluated on a real-world software defect dataset to improve recall for defective modules, supporting risk-based test automation.