Nashik · Open to opportunities

NileshSonawane.

Full stack developer shipping production CRM & SaaS, and the RAG systems that make them smart.

Scroll to assemble

01 / Interface

Interfaces people use every day.

React, Next.js and Tailwind front ends for CRM, HR, recruitment and quotation workflows, built to be fast and obvious.

02 / Core

One core that holds it all together.

Node, Express and MongoDB. Schemas, REST APIs, JWT auth and role-based access, owned from design to production debugging.

03 / Intelligence

Then I wire it into intelligence.

Multi-tenant RAG: crawl client sites, embed locally with Ollama, retrieve with Atlas Vector Search, answer with Gemini.

Interface → Core → Network

Here's what that looks like in production.

Loading frames 0%

Impact

Shipped. Measured.

Monthly Quotations

GST-compliant quotation engine driving master product data at Big Host India.

1,800s

down from 30 min

Paperwork Automation

Automated candidate offer/joining letters with instant email dispatch via Nodemailer.

Client Sites Live

Client websites served by the multi-tenant RAG chatbot, each with strict data isolation.

Pages Indexed

Chunked and embedded into per-tenant vector stores, kept fresh by scheduled re-crawls.

Selected Work

Built for production.

Three systems I designed and shipped, each one running for real users.

01/AI & Vector Search/2026

AI Website Knowledge Chatbot

Multi-tenant RAG Pipeline & Web Crawler

Engineered a production-ready multi-tenant RAG pipeline crawling customer websites, computing local embeddings with Ollama, storing vector representations in MongoDB Atlas, and generating grounded contextual responses via Google Gemini.

Dual-engine crawler combining fast Cheerio scraping with Playwright fallback for dynamic client-rendered JavaScript SPAs.
Local vector generation using Ollama and Nomic Embeddings to reduce third-party embedding latency and cloud cost.
Strict multi-tenant metadata partitioning inside MongoDB Atlas Vector Search preventing cross-tenant leakage.

Node.jsExpressMongoDB Atlas Vector SearchOllamaNomic EmbeddingsGemini APIPlaywrightCheerio

Pipeline

1Web Crawler (Cheerio/Playwright)
2Text Chunking
3Ollama (Nomic)
4Atlas Vector Search
5Gemini Grounded Answer

Outcome

10+ client websites, 100+ indexed pages, strict per-tenant data isolation

02/Full Stack SaaS/2026

Attendance Pro — Face-Verified SaaS

Cloud Attendance System with Computer Vision

Modern multi-role enterprise attendance platform featuring biometric face authentication via AWS Rekognition, GPS geofencing perimeter validation, and multi-channel WhatsApp OTP verification to eliminate proxy attendance.

Multi-role architecture supporting Super Admin, Company/Branch Admin, and Employee granular permissions.
Integrated AWS Rekognition facial comparison coupled with live browser camera streaming and liveness heuristics.
Tested self-hosted alternatives including InsightFace, MediaPipe, and ONNX anti-spoofing models.

React.jsViteTailwind CSSNode.jsExpressMongoDB AtlasAWS RekognitionGPS GeofencingWhatsApp API

Pipeline

1Camera Stream
2AWS Rekognition
3GPS Geofence Check
4Shift State Engine
5MongoDB Atlas Ledger

Outcome

Sub-second verification, zero proxy punch-ins, automated CSV/Excel reporting

03/Enterprise Systems/2026

Enterprise CRM & SaaS Suite

Production Business Management Platform

Architected core modules for Big Host India's production CRM, managing end-to-end recruitment pipelines, automated documents, lead conversions, employee lifecycle, and complex GST quotation calculations.

Designed normalized Mongoose schemas and high-throughput REST APIs for leads, recruitment, and branches.
Built one-click automated offer letter and onboarding contract generation with immediate Nodemailer delivery.
Formulated dynamic GST pricing logic driven by product master catalogs, handling multi-tiered tax calculations.

React.jsNode.jsExpressMongoDBMongooseJWT / RBACNodemailerTailwind CSS

Pipeline

1Role-Based React UI
2Express REST APIs
3Mongoose Models
4Nodemailer PDF Engine
5MongoDB Cluster

Outcome

1000+ GST quotes/month, paperwork time slashed from 30 mins to 1 sec

Experience

Where I build.

From cloud fundamentals to owning revenue-critical CRM modules.

May 2026 – PresentFull-Time

Software Developer

Big Host India Pvt. Ltd. · Nashik, Maharashtra

Building and shipping core modules for commercial CRM and SaaS products, driving automation, revenue-critical quotation engines, and security across the platform.

Architected Mongoose schemas, REST endpoints, and granular role-based permissions (RBAC) across CRM modules covering leads, recruitment, employee lifecycle, users, and multi-branch structures.
Automated offer letter and joining letter generation paired with instant transactional email delivery via Nodemailer, slashing paperwork turnaround from 30 minutes to 1 second per candidate.
Implemented high-precision, GST-compliant quotation calculation logic driven by product master data, reliably generating 1000+ quotes every month.
Spearheaded root-cause analysis and resolution of production issues involving CORS headers, routing bugs, JWT authentication tokens, and MongoDB replica set connectivity.

React.jsNode.jsExpress.jsMongoDBMongooseREST APIsRBACNodemailer

2023Internship

Cloud Intern

KasNet Pvt Ltd · Maharashtra, India

Hands-on exposure to public cloud infrastructure across AWS and Microsoft Azure.

Worked hands-on with core AWS and Microsoft Azure services.

AWSMicrosoft AzureCloud Infrastructure

Graduated 2024

Bachelor of Engineering (B.E.) in Computer Engineering

JSPM NTC, Savitribai Phule Pune University (SPPU)

Stack

Tools of the trade.

React.jsNext.jsTailwind CSSJavaScript (ES6+)Node.jsExpress.jsRESTful APIsMongoose & Schema DesignJWT & RBAC SecurityReact.jsNext.jsTailwind CSSJavaScript (ES6+)Node.jsExpress.jsRESTful APIsMongoose & Schema DesignJWT & RBAC SecurityReact.jsNext.jsTailwind CSSJavaScript (ES6+)Node.jsExpress.jsRESTful APIsMongoose & Schema DesignJWT & RBAC Security

RAG PipelinesMongoDB Atlas Vector SearchOllama & Nomic EmbeddingsGoogle Gemini APIAWS Rekognition (Face ID)MongoDBMongoDB Atlas (Vector Search)RAG PipelinesMongoDB Atlas Vector SearchOllama & Nomic EmbeddingsGoogle Gemini APIAWS Rekognition (Face ID)MongoDBMongoDB Atlas (Vector Search)RAG PipelinesMongoDB Atlas Vector SearchOllama & Nomic EmbeddingsGoogle Gemini APIAWS Rekognition (Face ID)MongoDBMongoDB Atlas (Vector Search)

AWS (Rekognition, S3, EC2)Microsoft AzurePostmanVS CodeCORS / Auth DebuggingNodemailer AutomationSpring Boot (Java)Hibernate / JPAAWS (Rekognition, S3, EC2)Microsoft AzurePostmanVS CodeCORS / Auth DebuggingNodemailer AutomationSpring Boot (Java)Hibernate / JPAAWS (Rekognition, S3, EC2)Microsoft AzurePostmanVS CodeCORS / Auth DebuggingNodemailer AutomationSpring Boot (Java)Hibernate / JPA

Frontend

React.js
Next.js
Vite
Tailwind CSS
JavaScript (ES6+)
CoreUI
Bootstrap
HTML5 / CSS3
Axios

Backend

Node.js
Express.js
RESTful APIs
Mongoose & Schema Design
JWT & RBAC Security
Nodemailer Automation
Spring Boot (Java)
Hibernate / JPA

AI & Automation

RAG Pipelines
MongoDB Atlas Vector Search
Ollama & Nomic Embeddings
Google Gemini API
AWS Rekognition (Face ID)
Playwright (Headless Crawler)
Cheerio (HTML Scraping)
Anti-Spoofing & ONNX

Data

MongoDB
MongoDB Atlas (Vector Search)
MySQL
Data Aggregations & Pipelines

Cloud & Tools

AWS (Rekognition, S3, EC2)
Microsoft Azure
Git & GitHub / GitLab
Postman
VS Code
CORS / Auth Debugging

Contact

Gotaproductthatneedstoship?Let'sbuilditproperly.

nileshsonawane5645@gmail.com

LinkedIn Call Resume