Aaradhya
Maharishi.
Software engineer, data & business analyst, and AI builder. I build test automation, AI-powered tools, and data-driven systems that turn messy requirements, processes, and data into clear, reliable, decision-ready operations.
About me
I'm a Computer Science graduate (JECRC University, CGPA 8.1) working as a Software Engineer at Capgemini, building test automation for a banking application. I work across software engineering, business and data analysis, operations, and AI — gathering requirements, documenting processes, writing the code, and measuring the result, so unstructured problems become clean, measurable systems.
I've shipped LLM-assisted analytics and automation across an internship at Deloitte Australia, and now bring that into live client-delivery engagements at Capgemini. Outside work, I'm a Top 20 Smart India Hackathon 2024 finalist and a core team member of Devcrest JU, mentoring a 230+ member coding community.
Experience
Software Engineer
// fortune 500 · consulting
- Engineered a TypeScript automation framework with Playwright and BDD (Cucumber) using the Page Object Model, replacing repetitive manual work with modular, reusable, repeatable runs.
- Translated functional requirements of a banking application into clear, structured workflows across UI, API, and hybrid (UI-to-API) flows, validating request/response accuracy across services.
- Shipped changes through a Jenkins CI/CD pipeline with Allure reporting, giving the whole team continuous visibility into build status and defects.
- Followed defined processes and checklists on live client-delivery engagements, applying AI-assisted tooling and documenting work for cross-functional teams.
Data Analytics & Forensic Technology Intern
// virtual internship
- Analyzed and processed 10,000+ records using Python (Pandas), Matplotlib, and Power BI — improving data accuracy by ~35% and cutting manual effort by ~40%.
- Built interactive Power BI dashboards with drill-down analysis to support forensic investigations and stakeholder decisions.
Projects
Noise2Signal
Domain-agnostic signal-discovery engine that turns noisy social chatter (YouTube, Reddit, or a CSV) into ranked, auditable ACCEPT/REJECT calls — scored on trend, sentiment, brand fit, and signal quality, with every score traced back to real comment text. Started as a single-purpose HealthKart flavor picker, now a reusable framework across categories — phone specs, sneaker colorways, or any custom domain — with zero changes to the scoring logic.
ParaBank Test Framework
End-to-end test automation for the ParaBank banking app — UI, API, and hybrid (UI→API) flows on the Page Object Model. Provisions unique users for safe parallel runs, cross-validates UI actions against live API balances, and ships Allure + HTML reports with RTM traceability, run through GitHub Actions and Jenkins.
Mocker — Exam Platform
Full-stack mock-exam platform with responsive UI, dynamic routing, and authentication. Analytics dashboards track performance and accuracy over time, with reusable components and clean data storage.
AI Code Review Bot
Autonomous CI/CD bot that triggers on every GitHub PR and posts AI-generated reviews instantly — using LLaMA 3.3 70B via the Groq API to flag bugs, security issues, and quality problems. Zero human intervention.
Mobile Sales Dashboard
Interactive dashboard exploring mobile-phone sales data — surfacing trends across brands, price segments, and regions with drill-down views built for quick, data-driven decisions.
Netflix Data Analysis
EDA across 6K+ Netflix titles by genre, year, rating, and region, with LLM-based summarization that auto-generates executive-ready insights — cutting manual write-up effort ~40%.
+ 13 more
18 public repositories in total — dashboards, experiments, and tools.
browse all repos ↗Skills & tools
languages
ai & automation
qa & testing
data & analytics
business & operations
visualization
databases & tools
certifications
Let's build something
worth shipping.
I'm open to freelance work, internships, and full-time roles in software engineering, data and business analysis, operations, and AI. If you have an interesting problem to solve, I'd love to hear about it.