Browse categories
Explore
Fiverr Pro
English
$
USD
I will build a full stack web app with intelligence and machine learning integration
Sohini B
About this gig
I build full-stack web applications where the frontend and the AI/ML layer are both done properly not a React shell calling a chatbot API and calling it "AI integration."
- On the frontend: React 19 or Next.js with Tailwind CSS, clean component structure, responsive across all screen sizes, fast load times.
- On the backend: Python with FastAPI, JWT-secured REST APIs, PostgreSQL or MongoDB, background scheduling where needed. The backend is production-ready, not a dev-server demo.
- For AI/ML: I can integrate LLMs (OpenAI, Groq, open-source models via API), build custom ML models (classification, anomaly detection, recommendation), and wire in real data pipelines not just a prompt wrapper.
Recent projects:
- QuantLedger: fintech portfolio tracker syncing live prices across 14 asset classes, with a quant risk engine (Sharpe ratio, volatility, cross-asset sensitivity matrix)
- NetSentinel: ML-based network intrusion detection system using XGBoost + MLflow + Evidently AI drift monitoring
- Spark: LLM-powered project visualization tool using Groq + Supabase + Next.js
Message me before ordering I'll tell you honestly whether your project fits my packages or needs a custom quote.
Get to know Sohini B
Sohini B
Full Stack Engineer Building AI Powered Apps
- FromIndia
- Member sinceJun 2025
- Avg. response time1 hour
Languages
Bengali, English, Hindi
Full-stack developer building products where React/Next.js meets real AI integration, not an API wrapper but actual ML pipelines underneath. Backend in Python/FastAPI, deploying ML models alongside REST APIs. Recent work –
QuantLedger: A fintech portfolio tracker with live market data and quant risk metrics.
NetSentinel: An ML-based intrusion detection system using XGBoost with full MLOps tracking (MLflow, Evidently AI).
Spark: An LLM-integrated project visualization tool.
I scope realistically, communicate clearly, and ship code that holds up past the demo, not just the pitch.
My Portfolio
Other Software Development Services I Offer
FAQ
What AI/ML features can you actually build — not just API calls?
Custom ML models (classification, anomaly detection, recommendations), LLM pipelines with real context/memory, model serving endpoints, and data drift monitoring. Not just ChatGPT wrappers.
What's your tech stack?
Frontend: React 19 or Next.js + Tailwind. Backend: Python + FastAPI + PostgreSQL. AI/ML: XGBoost, scikit-learn, OpenAI/Groq APIs, MLflow. Auth: JWT. Deployment: Docker.
Do I need to provide a design or will you create it?
I can work from your Figma file, a reference website, or a written brief. If you have no design at all, I'll suggest a clean layout — just describe what you need.
Will the code be mine after delivery?
Yes. Full source code ownership transfers on delivery. Delivered via GitHub repo or zip, your choice. No licensing restrictions.
Can you deploy the app after building it?
Basic deployment guidance is included. Assisted deployment to Vercel, Railway, or Render is available as an add-on — message me before ordering.
What if my project is bigger than the Premium package?
Message me before ordering. I do custom quotes for larger scopes. I'll give you an honest timeline and price — no mid-project surprises.
Do I need you provide you my API keys?
Yes, if it is a paid API service I will need your API keys. For free, open-source APIs, you can let me know if you have any specifics and I will make the necessary arrangements.
