Software Engineer
Kshitij Ijari
Building AI solutions that turn complexity into clarity.
Currently @ U-Haul · Phoenix, AZ
About
Bridging research
& production
I'm a software engineer who thrives at the intersection of academic innovation and practical engineering. With a Master's in Computational Science from SDSU and three published papers, I bring research-backed thinking to production systems.
Currently, I'm an SDET at U-Haul, building AI-powered tools that make QA and testing workflows more efficient. From self-healing test systems to MCP integrations, I create solutions used across SDET, QA, dev, and product engineering teams.
Career Journey
From startups to enterprise AI
Software Development Engineer in Test
U-Haul International
Phoenix, AZ · Dec 2024 - Present
Building AI-powered tools to streamline QA and SDET workflows. Designed custom indexers and self-healing test mechanisms with 85% success rate. Developing MCP tools for Azure DevOps used across QA, dev, and product engineering teams.
Software Engineer
Dori AI
Bangalore, India · Nov 2021 - Jun 2022
Led automated testing development reducing bugs by 35% and testing time by 40%. Built data pipelines and streamlined CI/CD with Docker, improving deployment times by 25%.
Full Stack Developer
PacketFanatic LLC
Florida, USA · Sep 2020 - Oct 2021
Led a team of 5 building scalable serverless financial apps. Reduced processing time from 4 hours to 4 seconds. Built Alexa-integrated AWS app achieving 50% faster order processing.
Featured Work
Impact at scale
Self-Healing AI Systems
U-Haul · AI Foundry
Engineered autonomous recovery mechanisms that detect, diagnose, and resolve production issues without human intervention.
Serverless Financial App
PacketFanatic · Team Lead
Led development of scalable serverless architecture processing financial transactions with automated testing frameworks.
QA Pipeline Automation
Dori AI · Software Engineer
Built comprehensive automated testing infrastructure with CI/CD integration, dramatically improving code quality and deployment velocity.
Alexa COVID Response
PacketFanatic · Integration Lead
Architected Alexa-integrated AWS serverless application addressing labor shortages during COVID-19 with voice-activated ordering.
Research
Published work
2D Image Reconstruction using Differentiable Plasticity
Optimized recurrent neural networks using Hebbian plastic connections for image reconstruction and memorization tasks.
Optimizing Recurrent Neural Network Using Neuromodulation
Improved model performance through neuromodulation and differentiable plasticity, achieving 3% accuracy increase with reduced training time.
Sustainable Pavement Management
Advanced pavement management system leveraging EfficientNetB3, ResNet, and GANs for road maintenance optimization.
Expertise
Technical toolkit
languages
frameworks
cloud
ai
tools
Get in Touch
Let's build something
together
I'm always open to discussing new opportunities, interesting projects, or just having a chat about AI and technology.