Experience the 3D Corridor: A Journey Through AI and Innovation
Welcome to Suyash Agrahari's Digital Universe
Welcome to the interactive 3D portfolio of Suyash Agrahari, a Software Engineer at HireQuotient specializing in autonomous AI agent development and full-stack engineering.
If you are searching for an innovative AI Engineer who builds intelligent systems with LangChain and LangGraph, a passionate Full Stack Developer who crafts scalable web applications, or an engineer who turns complex problems into elegant solutions, you have arrived at the perfect destination.
This website is not just a standard resume; it is an immersive, fully interactive WebGL experience built entirely from scratch.
When you enter this portfolio, you find yourself in a carefully hand-drawn 3D corridor. Every texture, every frame, and every interactive element was custom-crafted to demonstrate what is possible when cutting-edge AI technologies meet modern web development.
My goal as a Software Engineer is to build autonomous AI systems that solve real-world problems at scale.
At HireQuotient, I designed AI agents using LangGraph state machines that boosted matching accuracy by 65%, engineered multi-channel AI outreach platforms scaling to 10,000+ contacts, and optimized critical APIs from 29s to 1.5s (93% faster).
By utilizing powerful technologies like React.js, Next.js, Node.js, OpenAI SDK, and AWS, I deliver high-performance applications that drive measurable business impact.
The corridor serves as the central hub, offering doors that lead to different aspects of my professional journey: my AI projects, my technical expertise, and my contact information.
Explore my autonomous agent systems, featured builds like DrawFlow and SubhSandesh, and discover how I leverage AI to transform user experiences.
About Suyash Agrahari: The AI Engineer Behind Autonomous Systems
My Engineering Journey
My name is Suyash Agrahari, and I am a Software Engineer at HireQuotient with deep specialization in AI agent development and full-stack engineering.
My journey in software engineering began at The LNM Institute of Information Technology, where I earned my B.Tech (2020-2024), and evolved through building production-grade AI systems that drive real business impact.
At HireQuotient (May 2024 - Present), I design autonomous AI agents using LangGraph state machines and LangChain that intelligently match candidates with 65% better accuracy.
Previously, as a Full Stack Developer at Health Umbrella Foundation (Aug 2022 - Jul 2023), I built secure React.js and Node.js applications that boosted user engagement by 30%.
I believe that the future of software demands intelligent automation, where AI agents work autonomously to solve complex problems at scale.
Core Technical Skills & Expertise
As an AI-focused Software Engineer, my technology stack is built for creating intelligent, scalable, and high-performance systems.
- AI & Machine Learning: Expert in LangChain, LangGraph, OpenAI SDK, Google ADK, Hugging Face, RAG, Vector Databases (Pinecone, Chroma), and Guardrails AI. Built autonomous multi-agent systems using A2A protocol and agentic workflows.
- Agent Technologies: Specialized in building AI agents with tool calling, function calling, state machines, multi-turn coordination, and self-healing capabilities. Implemented intelligent systems for candidate matching, code review, and automated outreach.
- Core Programming: Advanced Python, JavaScript, TypeScript, C++, Java, with deep expertise in asynchronous programming, data structures, and algorithm optimization.
- Frontend Frameworks: Production-level expertise in React.js, Next.js, Redux Toolkit, Tailwind CSS, Material-UI, and Three.js for building interactive 3D web experiences with GSAP animations.
- Backend & APIs: Architected scalable Node.js and Express.js REST APIs, implemented WebSocket/Socket.io real-time communication, and built GraphQL endpoints. Expertise in RabbitMQ, Puppeteer automation, and n8n workflow orchestration.
- Cloud & DevOps: Deployed serverless functions on AWS Lambda, managed infrastructure on AWS EC2, S3, ECS, IAM, and API Gateway. Proficient in Docker containerization, Nginx configuration, and GitHub Actions CI/CD pipelines.
- Databases: Optimized MongoDB aggregation pipelines and indexes, implemented Redis caching strategies, and worked with PostgreSQL. Reduced API response times from 29s to 1.5s (93% faster) through database optimization.
The Gallery: Featured Builds and AI Projects
In the 3D gallery room, you can explore my most impactful AI systems, production applications, and open-source projects. Here is a detailed showcase of work created by Suyash Agrahari.
Autonomous AI Agent at HireQuotient - Production AI System
At HireQuotient, I designed and shipped an autonomous AI agent using LangGraph state machines and LangChain tool calling that intelligently matches candidates through iterative feedback loops—boosting matching accuracy by 65%.
I built an AI copilot agent with OpenAI SDK and function calling that lets recruiters automate sourcing, filtering, and outreach, cutting manual work by 70%.
Additionally, I engineered a multi-channel AI outreach platform using n8n, OpenAI/Gemini APIs, Node.js, and AWS EC2, scaling personalized campaigns to 10,000+ contacts and reducing workload by 75%.
This production system integrates 70+ AI tools via OpenAI API, Gemini API, and LangChain function calling, directly contributing to 35% growth in website traffic.
I also built an intelligent LinkedIn follow-up agent using Puppeteer and RabbitMQ that triggers personalized messages based on engagement behavior, improving reply rates by 50%.
DrawFlow - Open-Source Collaborative Diagramming Platform
DrawFlow (drawflow.in) is an open-source collaborative diagramming platform I founded and built as Solo Architect & Engineer.
Built with React, Node.js, WebRTC, and WebSockets, DrawFlow delivers Excalidraw's paid features entirely free.
I led the full product lifecycle from architecture to launch, drove organic growth through LinkedIn content marketing, and managed community engagement and user feedback loops.
This project showcases my ability to build real-time collaborative tools with complex state synchronization and peer-to-peer communication.
Visit the GitHub repository to explore the codebase.
SubhSandesh - Full-Stack Digital Celebration Platform
SubhSandesh (subhsandesh.in) is a full-stack digital celebration platform I founded and scaled to 500+ users.
Built with Next.js, Node.js, and MongoDB, I owned end-to-end product development, SEO strategy, and go-to-market execution, achieving 99.9% uptime and organic reach across 28+ Indian states.
This platform demonstrates my expertise in building scalable web applications with robust backend architecture, secure authentication, and optimized database queries.
SubhSandesh showcases how thoughtful engineering and strategic SEO can drive real user adoption and retention.
Autonomous Multi-Agent System (A2A Protocol) - Research Project
I architected a decentralized multi-agent system using A2A protocol + JSON-RPC 2.0 with capability discovery via /.well-known/agent.json for dynamic orchestration.
Built 10 specialized agents (Orchestrator, Blog Writer/Editor, Code Review, Landing Page Generator) using OpenAI function calling and LangChain BufferMemory for stateful multi-turn coordination.
Implemented a self-healing code review agent with iterative error detection for React hooks, Tailwind, and syntax issues, achieving 95%+ code quality improvement.
This project demonstrates cutting-edge AI agent architecture and inter-agent communication protocols. Explore the code on GitHub.
AI Interview Platform with Real-Time Proctoring
Built an AI interview system using OpenAI function calling and Kokoro TTS for JD/resume-based adaptive interviews with post-interview analytics and scoring.
Developed an anti-cheating proctoring system with face detection, tab monitoring, and copy-paste detection achieving 95%+ accuracy.
This platform showcases my expertise in building production-ready AI applications that combine natural language processing, computer vision, and real-time monitoring.
Perfect for companies looking to automate technical interviews with intelligent question generation and candidate evaluation.
Production Impact & Performance Optimization
At HireQuotient, I diagnosed and resolved a critical API bottleneck using MongoDB aggregation pipelines, Redis caching, and Promise.all(), cutting response time from 29s to 1.5s (93% faster).
I deployed a serverless YouTube media downloader on AWS Lambda supporting MP3/MP4/WAV formats that grew monthly traffic from 40K to 4.4M users (11,000% growth).
I delivered a full AI resume builder (Next.js, Redux Toolkit, Tailwind CSS) with 90+ ATS-optimized templates in just 20 days, speeding up product release by 60%.
Additionally, I recovered and re-published 3,000+ archived blogs through a Strapi CMS headless integration, improving SEO performance by 50%.
You can explore more projects and open-source contributions on my GitHub profile.