Abhinav Ranish

Software Engineer / Cybersecurity / AI

Abhinav Ranish

I like building useful systems end to end: AI tools with real data pipelines, security automation, browser extensions, research prototypes, and full-stack products that ship.

Technical Range

Frontend + Languages

ReactNext.jsTypeScriptPythonJavaC++CRustTailwind CSSAngular

Backend + Cloud

Node.jsDockerAWSAzureGCPPostgreSQLMongoDBSupabaseHeroku

Security + Tools

GitLinuxBashBurp SuiteMetasploitNmapWiresharkSplunkJira

AI / ML

PyTorchOllamaInsightFaceRAGFAISS

Experience

SWE Intern (Security + Frontend)

Ampcus Inc

Jun 2025 - Aug 2025
  • Built and presented an AI-powered Stock Analyzer using RAG architecture, sentiment analysis, and financial indicators—translating LLM outputs into actionable business insights for company leadership.
  • Shipped 10+ React components across 4 client modules; completed 150+ Burp Suite security labs and executed a live penetration test with authored findings report.

Technology Consultant (Part-time)

Enterprise Technology, Arizona State University

Nov 2024 - Present
  • Provided real-time technical support for faculty across AV, networking, and classroom systems; managed front-desk triage via Slack/phone/in-person and performed end-of-day readiness checks.
  • Performed end-of-day checks to confirm tech functionality and classroom security.

Information Technology Intern

Qatar Financial Centre Regulatory Authority (QFCRA)

May 2024 - Jul 2024
  • Provisioned 150+ employee endpoints with custom SOE Windows images; created network topology diagrams for leadership presentations during the organization's on-prem to Azure migration.
  • Created presentations on IT Infrastructure with tools like Trellix, Azure, and FortiClient.

Featured Projects

Selected for technical depth, shipped surface area, and practical engineering signal.

View all projects
IntelliTab preview

High impact / High complexity

IntelliTab

An AI-powered Chrome extension that intelligently organizes tabs into native Chromium tab groups, learns preferences over time, and manages workspaces. Features an adaptive learning system (passive snapshots, correction-based feedback, explicit preferences), an editable SOUL instruction set that guides AI grouping logic, workspace save/restore, auto-recovery across restarts, a rules engine for deterministic overrides, and multi-provider AI support (Groq, OpenAI, Gemini, Claude, OpenRouter).

  • Chrome extension using native tab groups and persistent workspaces
  • Multi-provider AI support with learning from user corrections
  • Rules engine for deterministic overrides
TypeScriptReact 18Tailwind CSSViteChrome Extension Manifest V3Multi-Provider AI
WiSpec preview

High impact / High complexity

WiSpec

Commodity dual-band Wi-Fi spectroscopy research for material classification and structural reconnaissance. Uses Wi-Fi signal reflections (RSSI and CSI) to characterize interior layouts, wall composition, and structural elements. Applications include building reconnaissance, search and rescue structural assessment, and smart building automation. Independent research by a Computer Science student at ASU.

  • Independent Wi-Fi spectroscopy research
  • Signal processing and material classification workflow
  • Built around commodity hardware and ML experimentation
PythonScapyPicoScenes802.11 Frame ParsingMachine LearningSignal Processing

Medium impact / High complexity

ClawHub

A lightweight, self-hosted GitHub clone optimized for autonomous agents. ClawHub provides a Git Smart HTTP server, API-first agent registration system, Personal Access Token authentication, and a web dashboard for repository management. Built with Node.js, Fastify, TypeScript, Prisma, Next.js, and PostgreSQL.

  • Git Smart HTTP server with API-first repository management
  • Agent registration and Personal Access Token authentication
  • Fastify, Prisma, Next.js, and PostgreSQL system design
Node.jsFastifyTypeScriptPrismaNext.jsPostgreSQL
FaceTagger preview

High impact / High complexity

FaceTagger

Offline AI tool to detect, embed, and cluster faces in 1TB+ photo libraries. UI built with Streamlit allows users to label clusters, write EXIF/JSON tags, and organize collections without cloud dependency. GPU acceleration via CUDA + InsightFace.

  • Offline clustering across large photo libraries
  • CUDA, InsightFace, Streamlit, and EXIF writing
  • Privacy-preserving local media workflow
PythonInsightFaceStreamlitEXIFToolONNXHDBSCAN

Open Source

Accepted upstream

Rust by Example Contributions

Authored 7 pull requests merged into the official Rust by Example repository, correcting examples and clarifying documentation used by Rust learners.

RustDocumentationOpen SourceGitHub

Merged PRs

7

Upstream Repo

rust-lang

Focus

Docs + examples

Research

Independent technical work with a clearer research story than a standard project card can hold.

WiSpec research preview
Independent student research

WiSpec

Commodity Wi-Fi spectroscopy research for material classification and structural reconnaissance. It combines wireless signal behavior, frame parsing, and machine learning to infer properties of indoor environments using off-the-shelf hardware.

3

Frequency bands

6

Material classes

<$200

Hardware cost

PythonScapyPicoScenes802.11 Frame ParsingMachine LearningSignal Processing

More Builds

A few more projects from the broader archive.

Open full archive

High impact

CVfy

An AI-powered portfolio generator that transforms resume uploads into fully functional personal websites. Built with Next.js (App Router), TypeScript, Prisma, and PostgreSQL. Features OAuth authentication, AI resume parsing via Groq/OpenRouter into strict JSON schema, 5 built-in themes, hero image uploads, slug-based publishing, and editable onboarding flow.

Next.jsTypeScriptPrismaPostgreSQLAuth.js

Medium impact

Velarith

Enterprise-grade AI financial analytics platform powered by Claude Sonnet 4.5. Built with Next.js, TypeScript, and Python Flask, Velarith delivers real-time AI market analysis with streaming responses, 15+ technical indicators, and multi-source data fusion from Finnhub, Yahoo Finance, and Polymarket. Created during the Claude Builder Club Hackathon 2025.

Next.jsTypeScriptPython FlaskClaude Sonnet 4.5Finnhub API

Medium impact

Posta

A full-featured Gmail enterprise clone for personal use. Built with Next.js 16, featuring OAuth authentication, email parsing and rendering, AWS S3 file storage with presigned URLs, email sending via AWS SES, SNS webhook processing, background job processing with Inngest, and a PostgreSQL database via Neon with Drizzle ORM.

Next.js 16TypeScriptDrizzle ORMNeon PostgreSQLAWS SES/S3/SNS

High impact

Net Audit Suite

Cyber audit tool integrating Nmap, Nikto, OpenVAS, and arp-scan to generate AI-powered summaries of discovered vulnerabilities using LLaMA + Ollama. Generates DOCX reports, maps infected IPs, and provides a dark-mode frontend in Next.js.

Next.jsNode.jsOllamaOpenVASNmap

High impact

Stock AI Analyzer

AI-powered stock analysis tool leveraging multiple data sources such as Google Trends, News API, and yfinance. It aggregates real-time market signals, sentiment, and price data to deliver actionable insights through a sleek frontend powered by AI.

PythonOllamaGoogle Trends APINews APIyfinance

Medium impact

Readflow

A modern, dark-mode-first Markdown editor built for speed, aesthetics, and instant sharing. Features a split-pane live preview with syntax highlighting, one-click generation of shareable read-only links, automatic local draft persistence, and Cloudflare D1 production-ready storage support.

Next.js 15Cloudflare D1@uiw/react-codemirrorCSS ModulesJavaScript

Let's build something useful.

Open to software engineering, cybersecurity, and AI opportunities where practical systems and clear execution matter.