Backend engineer at Wayground (formerly Quizizz) building microservices that process 100K+ events/day. I architect production systems in Go and Python, optimize database queries that handle millions of records, and design APIs that power platforms with 50K+ daily active users.

What sets me apart: I don't just write code—I ship features that scale. From reducing query times by 40% to building multi-agent AI systems from scratch, I obsess over performance, reliability, and clean architecture.

Current focus: Distributed systems, multi-agentic AI, and high-performance backend engineering.

• LeetCode: 1828 rating (Top 6.5% globally)
• CodeChef: 5⭐ rating
• Philosophy: Algorithmic problem-solving isn't just a hobby—it's how I approach system design, optimize database queries, and debug production issues.

The Problem: Creating high-quality podcast content takes time, planning, voice recording, editing, and distribution.
My Solution: Built a fully automated pipeline where users input a topic, and the system generates a complete podcast episode with scripts, narration, and voice synthesis.

Impact:

• Fully automated end-to-end workflow from topic → script → voice → downloadable audio
• Vector search–powered semantic retrieval for context-aware content generation
• Real-time progress tracking and async tasks for smooth user experience
• Scalable backend capable of handling concurrent podcast generation

Tech Stack: FastAPI, Next.js, PostgreSQL, Redis RQ, Qdrant Vector DB, HuggingFace, GCP TTS, AWS S3

Key Features:

• 🎤 Automated script generation using LLMs and semantic context
• 🧠 Vector search for topic-aware knowledge grounding
• ⚙️ Redis queue workers for async long-running processes
• 📡 Real-time UI feedback with streaming logs and progress indicators
• ☁️ Audio storage + delivery via AWS S3

Repo: NevraAI

The Problem: Most API gateways are either too complex or lack real-time observability.
My Solution: Built a high-performance distributed gateway with live analytics, request caching, and intelligent rate limiting.

Impact:

• Handles high-volume backend traffic with <10ms p50 latency
• Real-time SSE-powered analytics dashboard tracking p95 latency, cache hit ratio, and error rates
• Token-bucket rate limiting with per-key behavior
• Round-robin load balancing across backend instances

Tech Stack: Go, Chi Router, Redis, PostgreSQL (Neon), Next.js, Shadcn UI, TanStack Query

Key Features:

• ⚡ Sub-10ms reverse proxy with intelligent caching (configurable TTL)
• 🔐 API key authentication with tier-based traffic control
• 📊 Live metrics: latency distribution, traffic volume, failure patterns
• 🔁 Automatic failover and health checks

Repos: Backend | Frontend

The Problem: Most task tools lack true real-time collaboration and flexible role-driven workflows.
My Solution: Built a production-grade collaborative board system with live task updates, drag-and-drop workflow, and WebSocket-powered multi-user editing.

Impact:

• Real-time updates across users with WebSockets + Redis Pub/Sub
• Clean architecture with clear domains: controllers → services → repositories
• Fully client-side auth (NextAuth + JWT) with backend acting as a trust-based stateless API surface

Tech Stack: Go (Fiber), PostgreSQL, Redis, WebSockets, Next.js, Zustand, TailwindCSS, shadcn UI

Key Features:

• 🏗️ Drag-and-drop task board with multi-user state sync
• ⚡ Event-driven architecture with Redis Pub/Sub for broadcast operations
• 🔄 Real-time task editing, activity logs, and notification streams
• 🔐 Role-based UI enforcement (admin/editor/viewer) handled entirely on frontend
• 🧱 Clean modular backend architecture with services, repos, and DTO layers

Repos: Backend | Frontend

The Problem: Debugging API failures wastes hours. Staring at cryptic error messages in Postman isn't productive.
My Solution: Built a multi-agent AI system that diagnoses integration issues and suggests executable fixes in seconds.

Impact:

• Multi-agent reasoning isolates auth errors, schema mismatches, and network faults
• Generates actionable fixes with code snippets
• Secure per-session analysis with Clerk authentication

Tech Stack: Next.js, LangGraph, TypeScript, Clerk Auth, Groq LLMs, FastAPI

What makes it special:

• 🧩 LangGraph-powered multi-agent system for surgical fault isolation
• ⚡ Real-time diagnosis with streaming responses
• 🔐 Enterprise-grade security with session-based analysis
• 🧠 Learns from common API patterns to improve suggestions

Repos: Backend | Frontend

The Problem: Manual code reviews are time-consuming and miss architectural issues.
My Solution: Built an autonomous AI agent system powered by Groq's Llama 3.3 70B that performs comprehensive codebase analysis and PR reviews.

Impact:

• Scans entire repositories for architectural flaws, security vulnerabilities, and code quality issues
• Autonomous PR review with actionable recommendations
• Scales to 100+ concurrent requests with async FastAPI + Redis caching

Tech Stack: Python, FastAPI, Redis, Groq API (Llama 3.3 70B), Docker, Next.js, Shadcn UI

Architecture:

• Microservice 1: PR Review Agent – Autonomous pull request analysis
• Microservice 2: Codebase Analyzer – Full repository scanning with metrics
• Infrastructure: Dockerized deployment with CI/CD pipeline

Why it matters: Reduces code review time by 60% while catching issues human reviewers typically miss.

The Challenge: Build a production SaaS with payments, authentication, and 25+ productivity tools.
The Result: Full-stack platform with <200ms API responses, Stripe integration, and AI-powered features.

What I learned:

• Optimizing PostgreSQL queries for sub-200ms responses under load
• Integrating Stripe payments and webhook handling
• Why rate limiting isn't optional (learned this the expensive way)
• Smart caching strategies with Redis for 3x faster page loads

Tech Stack: Next.js, PostgreSQL, Stripe, GenAI APIs, Docker

Repo: Slanine

Performance Win: Integrated Redis caching to slash page load times by 42% (from 2.1s to 1.2s average).

Built a full-featured community platform with posts, voting, media uploads, and complete CRUD operations. The real achievement was architecting a caching layer that made the platform feel instant.

Tech Stack: Next.js, Redis, MongoDB, REST APIs

Repo: Qme

const rakshit = {
    backend: {
        primary: ["Go", "Python"],
        frameworks: ["FastAPI", "Chi Router", "Express.js"],
        architecture: ["Microservices", "REST APIs", "GraphQL", "Event-Driven Systems"]
    },
    databases: {
        sql: ["PostgreSQL", "MySQL"],
        nosql: ["MongoDB", "Redis", "Elasticsearch"],
        expertise: ["Query Optimization", "Indexing Strategies", "Caching Layers"]
    },
    frontend: ["Next.js", "React", "TypeScript", "Tailwind CSS", "Shadcn UI"],
    cloud: ["AWS (EC2, S3, Lambda, EKS)", "GCP (Compute Engine, BigQuery)"],
    devops: ["Docker", "CI/CD Pipelines", "Jenkins", "GitHub Actions"],
    ai: ["LangChain", "LangGraph", "Groq API", "OpenAI API", "Multi-Agent Systems"],
    
    currentlyMastering: [
        "Distributed Systems Design",
        "Multi-Agent AI Architectures", 
        "High-Performance Go Services",
        "Real-Time Data Processing"
    ]
};

Performance First: I don't just make things work—I make them fast. Every millisecond counts when you're serving 50K+ daily users.

Production Mindset: I build for scale. Error handling, observability, and graceful degradation aren't afterthoughts—they're requirements.

Learning by Shipping: Every project teaches something new. Whether it's distributed caching strategies or multi-agent AI systems, I learn by building real solutions.

Clean Architecture: Code should be self-documenting. If I can't understand it in 6 months, neither can my team.

I'm actively seeking roles where I can:

• Design and build distributed backend systems
• Architect solutions that scale to millions of users
• Work with modern tech stacks (Go, Python, AWS, Kubernetes)
• Collaborate with teams that value code quality and system design

Ideal roles: Backend Engineer | System Designer | Full-Stack Engineer (Backend-heavy)

Building something interesting? Let's talk.

• 💼 Portfolio: rakshitsisodiya.xyz
• 📧 Email: sisodiarakshit456@gmail.com
• 🔗 LinkedIn: rakshit-sisodiya
• 💻 GitHub: You're already here!

"The best code is no code at all. The second best is code that solves real problems efficiently."

⭐️ If you find my work interesting, consider dropping a star on repos that resonate with you!

_{Built with ❤️ by Rakshit | Last updated: November 2024}