About · Tucson, Arizona
Hi, I’m Parth.
I own the real-time data platform behind Komatsu’s autonomous haulage system — distributed Kotlin/Spring Boot services and AI-augmented developer tooling, including a custom MCP server that lets coding agents work with our APIs.
I’m a Software Engineer II in Tucson and the go-to technical liaison for the partner teams that build on the platform. Before that I shipped features to 200M+ users at Evernote and built deep learning pipelines at ISRO.
Always open to conversations about backend systems, AI tooling, or hard infrastructure problems —feel free to reach out.
The route so farFrom ocean eddies
From ocean eddies
to open-pit mines.
- Feb 2023 – now
Komatsu
Software Engineer II, autonomous haulage · Tucson- Own the Kotlin/Spring Boot data aggregation service that is the single source of truth for an autonomous haulage system — real-time telemetry from a fleet whose operations run $100K+/hour.
- Built an AI-augmented developer toolkit around a custom MCP server that lets coding agents search, call and validate our APIs, halving API verification time and doubling feature delivery.
- Cut developer onboarding by 75% and serve as the technical liaison for partner teams building on the platform.
2×faster feature delivery after building an AI dev toolkit around a custom MCP server. - Jun – Aug 2022
Evernote
Software Engineer Intern · Remote- Shipped “Repeat After Completion” recurring tasks to 50K+ beta users — time-zone-aware recurrence logic in TypeScript/GraphQL.
- Built a reusable task-filtering component that reached 200M+ users across web, mobile and desktop.
200M+users got the task-filtering component I shipped across web, mobile and desktop, in TypeScript and GraphQL. - Dec 2020 – Apr 2021
ISRO · Space Applications Centre
Research Intern · Ahmedabad- Built a Python deep learning pipeline (Mask R-CNN, YOLO) that found oceanic eddies in raw satellite imagery with 92% accuracy, batching terabytes of data without running out of memory.
92%detection accuracy teaching Mask R-CNN and YOLO to spot oceanic eddies in terabytes of satellite imagery.
Commit log
I build what excites me.
…and release it all as open source.