Headline

Frontend Engineer | React, TypeScript | Performance & Scalable UI | Rocketlane

About

I'm a frontend-focused engineer who builds high-performance React/TypeScript applications for B2B SaaS. I care about the metrics that make products feel fast. I've cut LCP from 5.6s to 1.8s, reduced heap memory by 250MB+ through smarter data fetching, and dropped app crash rates by ~90% by building out frontend reliability tooling. I also design and ship shared component systems (currently building one on Radix UI, ShadCN, and Storybook) that other engineers build on top of. I have working backend exposure — Spring Boot, Node.js, building and consuming APIs — enough to reason about the full stack when frontend performance depends on it, like designing a data pipeline that cut query time from 1.5s to 400ms. I like owning ambiguous problems end-to-end: breaking them into shippable milestones, instrumenting what "better" actually means, and mentoring through PR reviews and pairing along the way.

Location

Chennai

AI stack

antigravity

codex

Work experience

Software Engineer at Rocketlane, Aug 2023 - Present · 3 yrs 1 mo, Chennai, Tamil Nadu, India, - Architected Unified Dashboards with a custom drag-and-drop framework and widget virtualization — cut LCP from 5.6s to 1.8s via code-splitting, lazy loading, and deferred off-screen rendering. - Eliminated ~250MB of heap bloat (350MB → 100MB whole-tab) by replacing an eager bulk fetch with on-demand, debounced, batched requests backed by an LRU cache. - Drove a frontend reliability effort that cut production crashes from 1,000+/day to under 100/day (~90% reduction) — plus built a weekly agentic triage workflow (Claude querying Datadog/LogRocket) that autonomously opens fix PRs and flags anomalies to module owners. - Optimized the Kanban Board by replacing dnd-kit with native HTML5 Drag & Drop and task virtualization — DOM nodes 20k → 5k, heap 900MB → 300MB, drag latency 500ms → 42ms. - Building a unified design system (Radix UI, ShadCN, Storybook) used across the core platform and AI product Nitro, standardizing component primitives to counter AI-driven UI drift. - Designed and shipped Scheduled Exports end-to-end — a standalone TypeScript/Node microservice (SQS-driven, Puppeteer PDF rendering, S3 presigned URLs, K8s health probes) with full ownership of the architecture. - Architected Agile Planning modules (Sprint Planning, Epics, Backlog) in React and Spring Boot, including a Postgres → SQS → ClickHouse ingestion pipeline that cut query time from 1.5s to 400ms. - Set up Vitest and drove test coverage from 0 to 50%+, gated in CI; authored reusable Claude Code skills auto-applied during AI-assisted development for consistency.

Software Engineer at Rocketlane, May 2023 - Jul 2023 · 3 mos, Chennai, Tamil Nadu, India, • Enhanced the Project Management module by implementing features like inline edits and task rearrangements, improving user experience, and providing greater flexibility in managing project phases. • Developed and integrated Google SSO flow, enabling seamless user onboarding from the official website and simplifying authentication. • Revamped the login page, enhancing the design and user experience for smoother authentication. • Delivered multiple UI/UX design fixes, ensuring consistency and improving overall user satisfaction.

Education

Bachelor's degree, Mechanical Engineering, Manipal University Jaipur

Podar International School

Public links

github: https://github.com/partha8