Devansh!
ProjectsServicesAboutContactHire Me

Devansh!

Freelance Full-Stack & GenAI developer helping teams ship fast, reliable, and scalable AI products.

Navigate

HomeProjectsServicesContact

Reach out

GitHubLinkedInEmail

Focus

AI SaaS
Real-time Systems
Performance Engineering

© 2026 Devansh Baghel. Built with Next.js + Tailwind + intent.

Devansh Baghel • Full-Stack & GenAI Developer

I build high-performance AI & real-time apps

Full-Stack & GenAI developer who has shipped 3 production SaaS applications,

Hire MeLet's Build Together

Speed + reliability + product thinking.

Devansh profile photo
Available for work

What I bring to the table ?

  • Built 3 AI & real-time SaaS applications end-to-end
  • BTech @ NIT Bhopal
  • Solved 600+ DSA problems (strong problem-solving)
  • Experience with real-time systems & scalable backend
Book a Meeting

Currently open for

Software Engineer | Software Developer role

Also open to Freelance SaaS & GenAI builds

My Journey

1

XII (MP Board, PCMB)

2

JEE Mains'23 — Top 0.92%

3

NIT Bhopal — ECE'27

Skills

Tech stack I use to build fast and reliably

Organized by execution area so clients can quickly map capability to project needs.

Frontend

Next.jsReactTailwind CSSFramer MotionTypeScript

Backend

Node.jsExpressREST APIsWebSocketsSystem Design

AI

GenAI APIsPrompt WorkflowsModal GPU InferenceAI Feature Shipping

Tools & Cloud

MongoDBMySQLAWSVercelGitHub

Projects

Proof through real product builds

Each project is structured around business problem, engineering solution, and measurable impact.

Featured Build

AI Voice Studio

LiveCode

A voice-first AI SaaS to generate, edit, and publish studio-grade spoken content.

Problem

Creators needed a faster way to generate high-quality voice content without juggling multiple tools and slow rendering pipelines.

Solution

Designed a full-stack workflow with async AI inference queues, chunked processing, and a React-first editor for rapid iteration.

Impact

Cut end-to-end processing latency by up to 78% and improved project completion speed for iterative voice generation.

Metrics

  • 78% lower processing latency
  • Queue-based GPU inference pipeline
  • Production-grade auth and project workflows

Tech Stack

Next.jsNode.jsMongoDBGenAI APIsModal

Case Study

Opal AI

LiveCode

An end-to-end AI video SaaS platform where users can record, process, and share videos with built-in transcription and summarization.

Problem

Teams needed a single workflow to record, process, and distribute videos quickly without juggling separate tools or manual metadata work.

Solution

Built a full-stack SaaS pipeline with Next.js + Express.js, Socket.IO for real-time sync, AWS S3 + CloudFront for fast delivery

Impact

Delivered a fast video workflow with low-latency real-time collaboration and automated AI processing at production-ready scale.

Metrics

  • Reduced video load latency by 60% using CloudFront CDN
  • Automated ~90% metadata generation (5-min video processed in <30s)
  • Handled 20+ concurrent real-time rooms
  • Achieved <150ms event latency in real-time

Tech Stack

Next.jsExpress.jsSocket.IOAWS S3CloudFrontStripe

Case Study

Real-time Interview Platform

LiveCode

A collaborative interview environment with live coding, synchronized sessions, and instant interviewer feedback.

Problem

Remote technical interviews often lag, desync, and miss real-time collaboration context.

Solution

Implemented WebSocket-driven session state sync, real-time code events, and fault-tolerant reconnection strategies.

Impact

Delivered stable live collaboration under network fluctuations with lower event lag and smoother interviewer experience.

Metrics

  • Live session sync via WebSockets
  • Low-latency event handling
  • Resilient reconnection and room state recovery

Tech Stack

Next.jsWebSocketsNode.jsMongoDBVercel

Proof

Credibility at a glance

The numbers below reflect my focus on solving hard engineering problems with measurable business impact.

3+

Production SaaS Apps

Built and shipped for real users

78%

Latency Reduced

Optimized backend bottlenecks

600+

DSA Problems Solved

Strong problem-solving foundation

Real-time

Systems Expertise

WebSockets, events, and streaming

Services

What I can help you ship

Freelancer-focused execution across product engineering, AI integration, and performance architecture.

Full-Stack Web App Development

From product idea to production deployment with clean architecture and maintainable code.

  • Modern React/Next.js frontends
  • Scalable Node/Express backends
  • Production deployment + monitoring

AI-Powered SaaS Applications

I integrate GenAI capabilities into practical products that drive measurable user value.

  • Prompt + workflow orchestration
  • Inference pipelines for speed and cost
  • AI feature integration into existing apps

Real-Time Systems Engineering

Reliable, low-latency collaboration experiences using event-driven architecture.

  • WebSocket architecture design
  • Live updates, rooms, and syncing
  • State recovery and reconnection

Performance Optimization

Targeted audits and fixes that improve perceived speed, backend throughput, and UX quality.

  • Latency and bottleneck profiling
  • DB and API performance tuning
  • Frontend rendering optimization

About

Builder mindset with product ownership

I am Devansh Baghel, a 3rd-year ECE student at NIT Bhopal building full-stack and GenAI products with a strong focus on real-world outcomes.

I work best with startup teams that need fast execution, clear technical decisions, and production-grade quality from day one. I enjoy turning ambiguous product goals into concrete shipping plans and reliable systems.

I care about outcomes, not just code output. If a feature does not move the product forward, we rethink it.

I communicate clearly, ship in small reliable iterations, and keep systems understandable as they scale.

I bring a strong algorithmic base from solving 600+ DSA problems, which helps with backend design and optimization.

Contact

Let's build something impactful.

If you're building an AI product or a high-performance SaaS platform, I can help you ship faster without sacrificing code quality.

Primary channel

2311401105@stu.manit.ac.in
LinkedInGitHub
Start a Project