ATarpit.tiwari
Software Engineer · LevelUP MultiLink LLP

Arpit Tiwari

Software Engineer

I build backend systems, full-stack products, and AI-powered tools that solve real problems.

I build production systems across APIs, asynchronous processing, authentication, AI integrations, and workflow automation.

Backend EngineeringAI ApplicationsFull-Stack Development
View My WorkGitHubLinkedInContact

About Me

I'm Arpit Tiwari, a Software Engineer focused on building full-stack, backend, and AI-powered applications.

I enjoy turning ideas into reliable production systems, with hands-on experience across APIs, automation, authentication, asynchronous processing, and AI integrations.

At LevelUP MultiLink LLP, I've worked on AI-powered automation and internal intelligence products, including a prospecting platform and an investment intelligence dashboard with a retrieval-grounded AI assistant.

I'm always learning, building, and looking for meaningful engineering problems to solve.

Work Experience

Production engineering roles & key accomplishments

Junior Developer Associate

LevelUP MultiLink LLP
Feb 2026 – PresentRemote

Building production backend systems, full-stack products, and AI-powered automation across internal platforms.

Key Engineering Accomplishments

  • Integrated production APIs across prospect discovery, verification, enrichment, AI classification, and campaign generation.
  • Built async job processing with queues, caching, and rate limiting to handle high-volume workloads reliably.
  • Created 15–20 n8n automation workflows for lead generation and campaign execution.
  • Developed backend enrichment pipelines for an investment intelligence platform using Firecrawl and Ollama.
  • Implemented OAuth, JWT, RBAC, and Cloudflare Zero Trust for protected internal systems.
  • Containerized and deployed services using Docker and GitHub Actions CI/CD.
Core Technologies:FastAPIPythonNode.jsReactDockerOllamaClauden8nSupabase

Engineering Case Studies

In-depth architectural breakdowns of production backend, workflow automation, and AI systems built at LevelUP MultiLink LLP.

Professional Project · Source Code Proprietary · Built for LevelUP MultiLink LLP

AI-Powered ICP Prospecting Platform

End-to-End Prospect Discovery, Verification & Campaign Orchestration

An AI-powered prospecting platform that automates prospect discovery, verification, enrichment, and campaign generation.

The Problem

Manual prospecting required multiple disconnected processes for prospect discovery, email verification, classification, enrichment, outreach, and campaign creation.

How it works

1User ICP Input
2Prospect Search
3Email Verification
4AI Classification
5Lead Enrichment
6AI Outreach Generation
7Campaign Creation

What I built

Backend Architecture
API Orchestration
Async Processing & Job Queues
AI Integrations (Claude & Ollama)
Caching & Rate Limiting
Deployment Pipeline

Engineering details

  • Designed and implemented the backend orchestration pipeline integrating prospect search, email verification, AI classification, lead enrichment, and campaign generation across multiple third-party services.
  • Implemented asynchronous job queues with caching and rate limiting to support high-volume concurrent processing while preventing API contention and rate-limit violations.
  • Integrated Claude and Ollama to power AI-driven prospect classification and personalized outbound outreach generation.
Results

Automated prospecting, verification, enrichment, and campaign creation end-to-end, enabling generation of approximately 10,000–20,000 verified, ICP-matched leads per run without manual effort.

Tech stack

ReactFastAPIPythonDockern8nClaudeOllamaBlitz APIMillionVerifierSmartleadGetSalesGoogle Sheets
Professional Project · Source Code Proprietary · Built for LevelUP MultiLink LLP

Investment Intelligence Dashboard

Internal Intelligence Platform & Retrieval-Grounded AI Assistant

An internal investment intelligence platform that transforms WhatsApp-submitted pitch decks into structured company insights with an AI assistant powered by structured investment data.

The Problem

Investment research previously involved manual, per-deck research and information scattered across documents.

How it works

1WhatsApp Submission
2Document Intake
3Firecrawl Enrichment
4Structured Data Storage
5Ollama AI Assistant

What I built

Backend Enrichment Pipeline
Structured Data Layer
Authentication & RBAC Security
Ollama Integration
Retrieval-Grounded AI Assistant
API Credit Optimization
Retrieval-Grounded AI Assistant

Integrated Ollama to power a retrieval-grounded AI assistant that answers investment, portfolio, and deal-related questions using the platform's structured data rather than relying on generic LLM responses.

User Question
Structured Investment Data
Relevant Platform Context
Ollama LLM
Grounded Answer
💡 Engineering Constraint: The assistant is strictly constrained to the platform's verified structured investment data rather than relying on general model knowledge.
Security & Authentication Flow
Auth Pipeline:Google OAuth → JWT → Role-Based Access Control (RBAC) → Cloudflare Zero Trust
RBAC Roles:
AdminPartnerAnalystViewer

Engineering details

  • Built a backend enrichment pipeline that processes WhatsApp-submitted pitch decks and enriches company, founder, and funding data through Firecrawl.
  • Integrated Ollama to power a retrieval-grounded AI assistant using structured company data.
  • Implemented authentication and authorization using Google OAuth, JWT, role-based access control, and Cloudflare Zero Trust.
  • Reduced API credit consumption by caching Firecrawl responses and rate-limiting external calls.
Results

Replaced manual, per-deck research with an automated enrichment pipeline, giving management a centralized platform for investment intelligence, portfolio analysis, and AI-assisted deal insights.

Tech stack

ReactNode.jsWhatsApp BotFirecrawlSupabaseGoogle OAuthJWTCloudflare Zero TrustOllama

Open Source & Experiments

Public engineering projects are currently being prepared. Check GitHub for available repositories and experiments.

Explore GitHub

Technical Stack & Capabilities

Cleanly grouped technologies, tools, and backend engineering concepts used in production environments.

Backend

FastAPIPythonNode.jsExpressREST APIsJava

AI & LLM

ClaudeOllamaOpenAI APIGemini API

Infrastructure & DevOps

DockerGitHub ActionsCI/CDCloudflareVPS

Databases & Data

PostgreSQLMongoDBSupabase

Automation & Workflows

n8nFirecrawlAPI IntegrationsWorkflow Automation

Security & Auth

OAuthJWTRBACCloudflare Zero Trust

Frontend

ReactTypeScriptJavaScriptHTMLCSSTailwind CSS

Engineering Concepts

Queue ProcessingBackground JobsAsync ProcessingCachingRate Limiting

Coding Profile

Data Structures & Algorithms problem-solving background

LeetCode162+ Problems Solved

Solved across foundational topics:

Java (DSA)ArraysHashingLinked ListsQueuesPrefix SumsBinary SearchSortingBit Manipulation
View LeetCode →

Education

Indira Gandhi National Open University

Bachelor of Computer Applications (BCA)

Sep 2023 – Nov 2026

Let's build something useful.

Open to connecting with engineers, recruiters, founders, and teams working on software, backend engineering, and AI.

Email ArpitLinkedInGitHubLeetCode