Software Intern: Web Performance & Automation

Abhinay Kalkhanday

*********************@*****.*** 987-***-**** Portfolio LinkedIn GitHub

EDUCATION

Manipal University Jaipur Jaipur, Rajasthan

B.Tech in Computer Science (IoT & Intelligent Systems) Exp. Graduation: 2026 EXPERIENCE

Matrix Education Jaipur, Rajasthan

Software Engineering Intern Jan 2026 – Present

• Improved performance across 12 production websites, reducing load time from 45s to 5.6s, LCP from 9.6s to 0.9s, and TTFB from 4.0s to 0.3s.

• Supported platforms serving 31k monthly visitors and 388k requests/month, optimizing CDN delivery and caching using Cloudflare.

• Designed reusable CMS architectures and browser automation workflows using Playwright and Express.js across multiple production websites.

• Investigated and mitigated a production SEO spam incident affecting a live website by auditing database records, hardening deployments, and integrating automated dependency vulnerability monitoring. Intern ATS & SCADA Systems Jaipur, Rajasthan

Jaipur Metro Rail Corporation May 2024 – June 2024

• Analyzed complex ATS & SCADA architectures to author a technical training suite, subsequently adopted as the standard onboarding material for new interns.

PROJECTS

Content Operations Automation Platform GitHub Node.js, Express.js, Playwright, SQLite, PM2

• Designed a stateful job orchestration platform for long-running browser workloads using persistent queues, priority scheduling, and durable job storage.

• Implemented automatic crash recovery using a finite state machine supporting 8 worker states, 7 failure categories, and up to 5 recovery attempts, restoring service within 60 seconds after browser failures.

• Automated content maintenance workflows by processing stale or low-quality articles, identifying content issues, generating updated drafts, and routing rewritten posts to a human review queue before publication. Judicial Transparency Delay Analytics Platform GitHub FastAPI, PostgreSQL, Redis, Celery, Docker, Next.js

• Designed a judicial transparency platform that aggregates and analyzes public court data from 4 Indian judicial sources to identify delay patterns, adjournments, and case lifecycle trends.

• Built resilient ingestion pipelines with schema validation, anomaly detection, audit logging, and automated recovery to handle failures and source changes without data loss.

• Developed distributed background workflows using 29 Celery tasks, Redis, and scheduled retraining pipelines for ingestion, monitoring, and machine learning operations.

• Implemented machine learning models for case duration prediction, hearing outcome classification, and delay severity detection using historical judicial data.

Electra-Core: Intelligent Voter Roll Extraction Platform GitHub OpenCV, Tesseract OCR, Streamlit

• Designed a modular document intelligence architecture for extracting structured voter records from degraded voter roll PDFs using computer vision and OCR pipelines.

• Implemented a strategy-based extraction engine using 4 independent extraction strategies with automatic fallback handling for varying document layouts and scan quality.

• Built a quality-gated validation pipeline using 5 validation rules to detect extraction failures and automatically route low-confidence records for manual review.

• Developed a Human-In-The-Loop review system using Streamlit for correction workflows, audit reporting, and struc- tured Excel export generation.

• Processed 26-page voter roll documents containing up to 780 voter records in approximately 30–75 seconds while maintaining end-to-end traceability of extraction decisions. TECHNICAL SKILLS

Languages Java, Python, JavaScript, PHP, SQL

Backend FastAPI, Express.js, Node.js, REST APIs

Databases PostgreSQL, Redis, SQLite, MySQL, MongoDB Infrastructure Docker, Git, GitHub Actions, Cloudflare, Postman Automation Playwright, Browser Automation, Job Queues, Workflow Orchestration ML/AI Scikit-learn, spaCy, OpenCV, Tesseract OCR

Frontend HTML5, CSS3, Vanilla JavaScript

1

Contact this candidate