Online — Ready for Internship Opportunities

Shubham Bhattacharya

$ Backend Developer Intern

BCA Final Year student at Parul University (CGPA 8.13). Building resilient backend systems with Node.js, TypeScript, and PostgreSQL featuring agentic AI workflows, schema-validated tool calling, and production observability.

01.

Projects

Production-grade systems built to execute scale and resilience.

Gigi AI Ops Screenshot

Gigi AI Ops

v2.4.1-stable

Autonomous order-tracking and refund processing backend using custom ReAct loops, Zod schema protection, and Groq SDK multi-model fallbacks.

  • Built a custom ReAct orchestration engine (no LangChain) for autonomous order-tracking across multi-step reasoning cycles.
  • Designed a Zod-validated tool registry enforcing schema validation on every LLM output before database writes.
  • Implemented multi-model fallback routing (Llama 3.3-70B to Llama 3.1-8B) via Groq SDK for resilience during outages.
  • Integrated Sentry + Pino structured logging with tiktoken-based cost tracking for full-stack observability.
Node.js TypeScript Express Groq SDK PostgreSQL Zod Sentry
250k+ req/s <15ms latency
Medo Copilot Screenshot

Medo Copilot

v1.1.0-beta

AI launch-readiness auditor combining visual intelligence, Lighthouse performance metrics, and static code scans with automated LLM evaluations.

  • Built a 3-tier LLM waterfall (Gemini 2.0 → Groq/Llama Scout → OpenRouter/Claude) with automatic failover.
  • Designed a hybrid scoring engine combining AI audit reviews with deterministic Lighthouse and static metrics.
  • Developed automated screenshot capture for multimodal LLM visual regression reviews.
  • Built static security analyzers scanning for hardcoded secrets, code quality issues, and deployment risks.
Next.js TypeScript Vercel AI SDK Clerk Gemini Groq OpenRouter
99.999% Uptime 0 Data Loss
DocNow Screenshot

DocNow

v1.0.0-stable

Full-stack telemedicine SaaS platform offering patient triage classifiers, WebRTC consultations, and transaction billing structures.

  • Built a symptom triage classifier to route patients to appropriate specialists based on reported symptoms.
  • Developed WebRTC video consultations via Vonage API — browser-based, no downloads required.
  • Implemented credit system with full audit trail: purchases, deductions, refunds, and transaction history.
  • Built multi-role dashboards (Patients, Doctors, Admins) with automated payout engine and fee calculations.
Next.js TypeScript Clerk Prisma PostgreSQL Vonage WebRTC shadcn/ui
50M+ Docs <5ms search
02.

Tech Stack

Technologies and toolkits I use to architect and support production backend systems.

Languages

TypeScript
JavaScript (Node.js)
SQL

Backend & API

Express.js
Next.js
REST API Design
Middleware
SSE / WebSockets
Rate Limiting
Auth (Clerk)

AI & LLM Infrastructure

Custom ReAct Loops
Tool Registry Design
Multi-Model Routing
Prompt Injection Shield
tiktoken Estimation
LLM Observability

Databases

PostgreSQL
Prisma ORM

Observability & Safety

Zod Schema Validation
Sentry Error Tracking
Pino Structured Logging

CI/CD & Infrastructure

GitHub Actions
Docker Compose
Vercel
03.

About & Core Principles

Shubham Bhattacharya

Shubham Bhattacharya

BCA Final Year — Parul University | Vadodara, IN

My engineering philosophy is rooted in architectural minimalism. I believe the best code is code that doesn't need to be written, and the best systems are those that are trivially easy to reason about when things go wrong under high workloads.

I focus primarily on backend API designs, distributed task layers, database performance, and production observability. Currently looking for Backend Developer Internship roles where I can contribute to core server architectures and database layers.

04.

Establish Connection

Currently seeking Backend Developer Internship opportunities to contribute to REST APIs, database architecture, and backend reliability. Reach out through any of the channels below to discuss opportunities or technical collaborations.

Email Me Resume