Backend Developer @ Equitysoft Technologies

Dhaval GondaliyaBackend Developer

I build the part of the product nobody sees — and everyone depends on.

Backend engineer shipping production APIs, payment flows, real-time systems and AI pipelines with Node.js, NestJS, PostgreSQL, MongoDB and AWS. I own features from schema design to the dashboards that tell us they're healthy.

See the systems →Download résuméEmail me
2+
years shipping production backends
18
client API codebases worked across
8+
third-party platforms integrated
3
payment gateways taken to production
01 — selected systems

Five systems, five different hard problems.

Each one had a failure mode that mattered — double payouts, leaked tenant data, slow AI calls, million-row imports. Here's what I built and how the request actually flows.

AI · RAG · Retail

Sorto

AI-powered wardrobe & closet management

the problem

Users add clothes from a photo or a shop link and expect the app to know the brand, category and attributes — and to suggest what matches.

Photo / URL
Scraper + S3
BullMQ worker
Embeddings + GPT
pgvector / Pinecone
Recommendations
  • RAG pipeline that classifies items with HuggingFace embeddings and GPT, using engineered prompts to extract brand, category and attributes.
  • Vector similarity search on pgvector and Pinecone to find similar wardrobe and catalog items for recommendations.
  • Web-scraping pipeline (Puppeteer, Cheerio) enriches items with catalog metadata used as retrieval context.
  • PostgreSQL triggers and stored procedures automate wardrobe processing; an “aura score” gamification layer drives engagement.
Node.jsPostgreSQLpgvectorPineconeHuggingFaceOpenAIBullMQRedisAWS S3Puppeteer
02 — also shipped

More production APIs I've worked on.

Client products at Equitysoft across commerce, health, sports, events and social — each with its own integrations, background jobs and deployment on real infrastructure.

Meastoria

Commerce & credits platform

Credit system, Stripe payments, analytics APIs, multilingual invoices and PDF generation, with a Next.js web app and React admin.

StripeSocket.ioPDFi18n

TipRadar

Sports data & predictions

Ingests live sports feeds (SportMonks, basketball APIs) through scheduled workflows, cached in Redis and served in multiple languages.

RedisCronSports APIsi18n

Quick Auth

Authentication service

JWT sessions, TOTP two-factor auth with QR enrolment, SMS OTP over Twilio and Stripe-backed plans.

JWTTOTP 2FATwilioStripe

Powerhouse

Operations & communications

RingCentral telephony, real-time socket workflows, and generated PDF and Excel documents via Puppeteer and pdf-lib, with Swagger docs.

RingCentralSocket.ioPuppeteerSwagger

Xcelable

Events & conferences

QR-based conference check-in, Twilio messaging, presigned S3 uploads and server-rendered PDF badges, localised.

QRAWS S3TwilioPuppeteer

Patient Wellness

Healthcare screening

Vision-screening platform with generated PDF reports, Excel exports and push notifications; Next.js app and React admin.

PuppeteerFirebaseExcelHealth

DanceSpot

Community & bookings

Stripe payments, Twilio messaging, real-time chat over Socket.io and scheduled jobs.

StripeTwilioSocket.ioCron

FitNext

AI fitness coaching

OpenAI-assisted fitness features with a unified statistics API across workouts and progress.

OpenAIAnalyticsCron

TasteWise

Food discovery

Gamified badge progression, advanced sourcing filters and Swagger-documented APIs.

GamificationSwaggerTwilio

NewMe

Nutrition tracking

Nutrition data from the USDA FoodData API, normalised into a searchable food database.

USDA APIMySQLCron

CircleUp

Social circles

Real-time chat, deep links into the Flutter app and scheduled engagement jobs.

Socket.ioDeep linksCron

The Lakewood Luach

Community calendar

Excel-driven data imports verified with Jest and property-based tests (fast-check).

Jestfast-checkExcelJS
03 — how I build

Six rules I don't ship without.

The difference between an API that works in Postman and one that survives production is mostly decisions like these.

  1. 01

    Money moves exactly once

    Payouts, refunds and prize distribution run inside transactions with idempotent writes, so a retry can never pay someone twice.

    ↳ StayFit prize engine · SpaceRenta host payouts
  2. 02

    Tenants never meet

    When a client's data is the product, isolation beats cleverness — a database per tenant, with a registry that knows where everyone lives.

    ↳ PlayerScout multi-tenant SaaS
  3. 03

    Slow work belongs in a queue

    Scraping, AI classification, PDF rendering and bulk imports go to background workers, so the API answers in milliseconds and jobs retry safely.

    ↳ BullMQ + Redis in Sorto · streaming CSV imports in Eterna
  4. 04

    A slow query is a bug

    Indexes, aggregation pipelines and caching are part of the feature, not a follow-up ticket. I read query plans before I add servers.

    ↳ Query + index restructuring across live Equitysoft APIs
  5. 05

    Configuration over deployment

    Roles, permissions and business rules live in the database so a client can change them on Tuesday without a release on Wednesday.

    ↳ Dynamic, database-driven RBAC
  6. 06

    If it isn't logged, it didn't happen

    Structured logs, monitoring and alerts ship with the feature, so production issues are caught before users report them.

    ↳ PM2 + CloudWatch monitoring on AWS
04 — toolbox

What I reach for, and what I've plugged into.

Core

Node.jsNestJSExpress.jsTypeScriptJavaScriptPythonSQL

Data

PostgreSQLMongoDBMySQLRedispgvectorPineconePrismaTypeORMMongoose

Architecture

REST API designMulti-tenancyRBACWebSocketsQueues (BullMQ)Cron jobsCachingRAG pipelines

Cloud & Ops

AWS EC2S3RDSLambdaSQSCloudWatchRoute53NginxPM2GitHub ActionsLinux

Also comfortable with

React.jsNext.jsDjangoFastAPISwagger / OpenAPIPostmanJest

integrations in production

StripeRazorpayPaystackTwilioRingCentralFirebase FCMOpenAIHuggingFacePineconeGoogle APIsAWS S3AWS RDSAWS SQSSportMonksUSDA FoodDataNodemailer
05 — experience

Where I've shipped.

Backend Developer

Equitysoft Technologies

Apr 2025 — Present
  • Own end-to-end backend development for production systems on Node.js, MongoDB and SQL: APIs, background jobs, integrations and error handling for live workloads.
  • Restructured database queries and indexes to cut API response times and improve reliability.
  • Run deployments, server maintenance and AWS scaling for production uptime under real traffic.
  • Set up monitoring and structured logging to catch production issues before users do.
  • Plan, estimate and ship features with frontend, QA and product teams in Agile sprints.

Node.js Developer

Thinknovus Technologies

Dec 2023 — Apr 2025
  • Built and maintained REST APIs on Node.js and MongoDB with authentication, WebSocket features, payments and caching.
  • Designed cron-based background jobs for scheduled notifications and data syncs.
  • Improved performance through query tuning, Redis caching and better schema design.
  • Managed CI/CD pipelines, deployments and infrastructure monitoring on AWS.
2021 — 2025

B.E. Computer Engineering

LJIET, Ahmedabad · CGPA 8.6 / 10

2021

Higher Secondary (Science)

GSEB, Ahmedabad · 86.66% · 87.83 percentile

06 — contact

Need a backend that holds up in production?

Open to backend and platform engineering roles and interesting freelance builds. I usually reply within a day.

+91 9586482626LinkedIn ↗GitHub ↗