Refactored a monolithic repo into a modular code structure, clarifying boundaries, reducing coupling, and enabling faster, safer perf-focused changes.
Instrumented application and browser signals in Datadog & New Relic to watch CWV-adjacent triggers (slow routes, error spikes, long tasks), with dashboards and alerts for perf degradations.
Regrouped and standardized Postman collections (environments, variables, naming), improving maintainability and enabling automated smoke/regression/performance runs in CI.
Built automated smoke, regression, and performance tests, giving the team fast feedback on hot paths; documented runbooks and failure triage.
Logged and tracked defects in Zoho Sprints, tightening the loop from alert→repro→fix→verify.
Partnered with devs and PM to define “fast-by-default” acceptance checks for new features (basic render time, payload sanity, third-party constraints).
Built a real-time SER pipeline (streaming audio → feature extraction → CNN/LSTM inference) with sub-second latency and 80%+ accuracy on internal validation.
Exposed inference via a FastAPI service (REST + WebSocket streaming) and a React demo UI; added client-side performance budgets and lazy loading for fast startup.
Designed multimodal fusion path (audio + visual cues) and published a journal on multimodal SER; prototyped synchronization and late-fusion strategies.
Implemented CI with unit/integration tests and performance gates (baseline latency/throughput) in GitHub Actions; reproducible builds with Docker.
Instrumented observability (timers, percentiles, error rates) and ran load/soak tests with JMeter and Postman collections to prevent regressions.
Shipped edge-ready packaging (ONNX/TFLite) to enable lightweight deployment and data-privacy-friendly local inference.
Drove web performance QA for the admin console and web surfaces, auditing with Lighthouse and Chrome DevTools (Performance/Network/Memory) to spot rendering, JS execution, and payload bottlenecks.
Introduced Core Web Vitals tracking (LCP, INP, FCP) and baseline budgets; added Lighthouse CI checks in GitHub Actions to flag regressions pre-merge.
Partnered with frontend to validate code-splitting, lazy loading, and preloading; verified bundle size budgets and impact of third-party scripts.
Set up lightweight synthetic monitoring and RUM sampling to correlate CI signals with production behavior; created perf triage runbooks.
Verified caching headers, image/font optimization, and critical-path strategies; ensured SEO and performance goals were aligned on public web pages.
Used React Profiler and DevTools flamecharts to identify re-render hotspots and long tasks; filed actionable issues with acceptance criteria tied to CWV targets.
Led QA for WAVEME, a mobile-first AI super app, ensuring performance, security, and cross-platform consistency across Android and iOS.
Developed and executed 300+ test cases (functional, UI/UX, integration, regression, edge), driving a 45% drop in post-release defects.
Built reusable automated test scripts in Selenium (Python) and integrated into GitHub Actions CI/CD, cutting manual testing time by 60%.
Conducted API testing with Postman and REST-assured, validating 120+ endpoints (chat, posts, notifications, auth).
Collaborated with backend teams (Rust, Node.js) to resolve performance bottlenecks, improving API response times by 30%.
Ran load and stress testing via JMeter for real-time messaging/media sharing, ensuring stability under 5,000+ concurrent users.
Participated in Agile ceremonies (standups, sprint planning, retros), boosting sprint testing velocity by 30%.
Reported and tracked 180+ bugs in Jira with clear repro steps, severity, and evidence for efficient dev handoff.
Assisted testing of decentralized messaging components using IPFS and LoRa, validating offline-first and censorship-resistant behavior.
Maintained QA documentation and a Confluence knowledge base, improving onboarding and test reusability.
Partnered with designers to ensure UI consistency and usability, reducing customer-reported UI issues by 35%.
Spearheaded quality assurance and system validation for Summit AI, an intelligent document processing platform leveraging OCR, LLMs, and cloud-based AI services.
Developed and maintained backend modules in Python, integrating Google Cloud Vision, Tesseract, and OpenAI for smart document classification and extraction.
Built automated test pipelines using PyTest, Django Test Framework, and Postman, achieving 80%+ test coverage across OCR and upload workflows.
Designed robust scenarios for AI model behavior validation, edge cases, and asynchronous background task execution.
Conducted end-to-end performance audits across the document pipeline, removing latency in PDF parsing, OCR tokenization, and LLM output generation.
Refactored legacy synchronous processes to asyncio + Dramatiq, reducing large-document response times on heavy workflows.
Implemented regression, integration, and usability tests for multi-page, rotated, and noisy documents to harden AI reliability.
Collaborated with data scientists to tune model parameters, track inference errors, and design fallback mechanisms.
Led QA sign-offs for releases, verifying OCR precision, JSON integrity, quota enforcement, and overall UX.
Built CI/CD workflows in GitHub Actions with Docker for automated testing, linting, and staging deployments.
Authored internal QA and AI evaluation docs and reporting templates for feature launches and performance benchmarking.
Contributed to refinement of prompts and pre/post-processing to improve relevance, consistency, and hallucination control.
Built custom ERP platforms using React.js, Node.js, and Firebase, streamlining inventory, HR, and sales workflows.
Deployed applications on Vercel for high uptime and fast global delivery.
Integrated AI-enhanced features for analytics dashboards and smart search.
Ensured cross-device compatibility, mobile responsiveness, and strong SEO fundamentals.
Worked with CMS platforms like WordPress, Joomla, and Squarespace.
Developed custom admin panels with real-time reporting and role-based access controls.
Implemented RESTful APIs for seamless integration with external services (e.g., accounting and logistics tools).
Led UI/UX design sprints to deliver intuitive, user-friendly interfaces for non-technical staff.
Created reusable components and modular architecture for future scalability.
Conducted internal training sessions for staff on managing and updating ERP modules.
Optimized ERP load times through frontend bundling and API response tuning.
Wrote technical documentation and user manuals for post-deployment handover.
Spearheaded web performance across multiple client platforms, delivering load-time reductions up to 60% and significantly improving Core Web Vitals (LCP, INP, FCP).
Ran in-depth audits with Lighthouse, WebPageTest, and Chrome DevTools, identifying bottlenecks in rendering, JS execution, and network payloads.
Implemented code splitting, lazy loading, tree shaking, plus font and image optimization.
Partnered with product teams to rework performance-critical flows, improving perceived speed and engagement on high-traffic pages.
Introduced critical-path optimization and SSR enhancements in Next.js apps deployed on Vercel.
Built custom performance dashboards and real-time telemetry using New Relic, Firebase Performance, and custom Node.js monitors.
Reduced third-party script impact via async loading strategies, sandboxed execution, and graceful fallbacks.
Optimized server response times by refactoring Node.js logic, tuning PostgreSQL queries, and adjusting Cache-Control headers.
Aligned SEO with performance goals, increasing organic visibility.
Performed A/B testing and synthetic monitoring to validate improvements and prevent regressions.
Led internal training workshops on web performance best practices for engineers, QA, and PMs.
Platforms & builds: Webpack 5, Vite, Rollup; code-splitting, preloading, bundle budgets.
Led development of high-performance web platforms using Next.js, React, and TypeScript, focusing on clean architecture, SSR, and SEO optimization.
Designed and implemented scalable backend services using Node.js, Python (FastAPI/Django), and Scala, enabling robust data processing and API integration.
Collaborated with AI researchers to integrate OpenAI, Hugging Face, and custom ML models into customer-facing interfaces for smart search, personalization, and prediction.
Architected serverless workflows using Firebase and Google Cloud Functions, ensuring rapid iteration and minimal infrastructure overhead.
Developed data pipelines for ingesting and analyzing user behavior, integrating BigQuery, Firebase Analytics, and custom ETL processes in Python.
Refactored and modularized large legacy codebases, improving code readability and reducing technical debt by 35%.
Optimized front-end performance, achieving top Core Web Vitals via code-splitting, lazy loading, and asset compression strategies.
Mentored junior developers through pair programming, code reviews, and architecture planning, establishing best practices across frontend and backend teams.
Worked closely with product, design, and QA to deliver sprint-based releases with high velocity and quality.
Implemented multi-tenant architecture support and role-based access control (RBAC) for enterprise-grade applications.
Deployed production apps using Vercel, Docker, and CI/CD in GitHub Actions, ensuring zero-downtime rollouts.
Contributed to internal developer tooling, including reusable UI libraries, testing utilities, and performance dashboards.
Supported the rollout and maintenance of Microsoft Dynamics 365 for finance, procurement, inventory, and HR departments across the enterprise.
Collaborated with consultants and internal stakeholders to tailor D365 modules to organizational processes, including vendor management, payroll, and stock control.
Assisted in user training and onboarding by creating guides, hosting walkthrough sessions, and providing one-on-one support.
Participated in business process mapping, documenting workflows and pain points prior to system customization.
Configured user roles, access levels, and security profiles in line with internal audit and compliance requirements.
Helped track and resolve system issues during UAT and post-deployment phases, documenting bugs and coordinating with support teams.
Conducted data entry, cleanup, and migration to ensure smooth transition from legacy spreadsheets and manual systems to D365.
Tested module customizations and updates to ensure accurate financial and operational reporting.
Contributed to system update rollouts, validating integrity across purchase, sales, and general ledger modules.
Participated in post-implementation reviews, offering feedback on system usability and areas for future automation.
Senior Web Developer & Frontend Performance Engineer specializing in React/Next.js and platform-level guardrails.
I set and enforce Core Web Vitals targets (LCP/INP/CLS), wire Lighthouse CI into PRs, and use Chrome DevTools/React Profiler to eliminate re-renders, long tasks, and memory leaks.
Strong with SSR/streaming/hydration, bundle/caching strategy (Webpack 5, code-splitting, preloading), and observability (Sentry, Datadog, New Relic).
I partner across product/design/QA to make fast-by-default the default while serving Quality.
Discover other experts with similar qualifications and experience
2025 © FRATCH.IO GmbH. All rights reserved.