Ajinkya Kunjir
Software Developer in Test, SDET and computer science researcher
Thunder Bay, Ontario, Canada
Joined February 2026
Network
1.3K connectionsMSSPMJ
JCDCTF
JWTVEZ
SPCTMJ
MDFGCE
PPRV
KBVN
Summary
Research-oriented computer scientist with a focus on applied machine learning and source-code analysis. During graduate studies and subsequent publications, he worked on code-similarity and plagiarism-detection approaches using lexical analysis, tokenization, and supervised learning methods. gvpress+2
Practitioner in software testing and SDET roles with experience across game testing and enterprise QA; writes and researches about improving testing workflows and accessibility in AI-generated UI. theorg+2
Technical communicator who publishes practical guides on prompt engineering and AI tooling, aiming to make advanced prompting patterns accessible to practitioners. testaholicsanonymous+1
Academic and teaching experience: completed a Master of Science in Computer Science and served in lecturing roles, combining classroom instruction with research and publication activity. mecs-press+1
Work
Education
Projects
Writing
Prompt Engineering Guides: Google vs. OpenAI 3 What You Actually Need to Know
May 1, 2025A distilled guide comparing prompt engineering practices for Gemini/Vertex AI and OpenAI/GPT, with practical tips and presets for model settings.
A HYBRID APPROACH TOWARDS LOCALIZATION AND SECURITY IN WIRELESS NETWORKS
November 1, 2018Conference/journal article on localization and security in wireless networks (co-authored; full text available on ResearchGate).
Your AI Is Copying Bad Code. Here’s How To Get It To Stop.
Article describing a controlled experiment on prompting AI coding tools to produce accessible, WCAG-compliant UI; advocates a native-first prompting strategy to improve accessibility outcomes.
Developing Machine Learning Coding Similarity Indicators for C and C++ Corpuses
Research on building a forensic similarity detection engine for source-code plagiarism detection using lexical analysis, tokenization, and supervised learning for C/C++ corpuses.