Hello
LP.
Graph
FOUNDING ENGINEER / AI / FULL STACK

I build
things that work.

I build AI-powered products from 0 → 1, owning architecture, engineering, AI integration, and production delivery.

FocusAI · Product · Systems
RoleFounding Engineer
CurrentlyBuilding Graph
ElsewhereGitHub·LinkedIn·Graph·Estana
01 / SELECTED WORK

Engineered Evidence

01 / 07[ PRODUCTION ]
mygraph.idCase Study

Graph

An event-driven professional identity platform for Data, AI, and Analytics professionals. Built around dynamic career timelines, evidence-backed professional profiles, and AI-powered skill enrichment.

Verified Metrics & Impact
  • -Founding Engineer
  • -Event-Driven Professional Identity Engine
  • -3-Horizon Timeline (Today, Event-Driven, Over-Time)
RoleFounding Engineer
Tech StackNext.js · TypeScript · RAG Embeddings · Vector Search · Event-Driven Arch
02 / 07[ PRODUCTION ]
estana.inCase Study

Estana

An AI-powered, multi-tenant recruitment platform built for recruitment agencies. Designed and shipped the platform end-to-end, from database architecture and candidate matching to AI-powered resume parsing and production infrastructure.

Verified Metrics & Impact
  • -3,000+ candidates processed in 1.5 months
  • -RAG & Few-Shot Prompts for 98% Parsing Precision
  • -Sole Architect & Engineer from Schema to Infra
RoleFounding Engineer
Tech StackNext.js · Node.js · PostgreSQL · Prisma · RAG Pipelines
03 / 07[ ACTIVE SYSTEM ]
mybotgenie.aiCase Study

MyBotGenie AI

Production RAG assistant that ingests company documents and answers questions with source citations. Custom retrieval pipeline with vector search, chunking strategies, and streaming LLM responses.

Verified Metrics & Impact
  • -10K+ documents indexed
  • -<800ms time-to-first-token
  • -95% citation accuracy
RoleSole AI Architect
Tech StackNext.js · OpenAI · Pinecone · LangChain · Tailwind
04 / 07[ ACTIVE SYSTEM ]
Case Study

AI Interview Analyst

End-to-end system that transcribes interview recordings, extracts key insights, and generates structured analysis reports. Audio processing, speaker diarization, and LLM-powered summarization.

Verified Metrics & Impact
  • -60min audio → report in <2min
  • -90%+ speaker diarization accuracy
  • -Structured scoring across 12 dimensions
RoleSole AI Architect
Tech StackPython · Whisper · GPT-4 · FastAPI · React
05 / 07[ ACTIVE SYSTEM ]
Case Study

Semantic Search Engine

Document search platform converting unstructured text into vector embeddings with contextually relevant retrieval. Multi-tenant indexing, hybrid keyword + semantic search, and relevance tuning.

Verified Metrics & Impact
  • -100K+ documents indexed
  • -<500ms p95 retrieval latency
  • -30% relevance lift over keyword-only
RoleSole AI Architect
Tech StackOpenAI Embeddings · Pinecone · Next.js · Node.js · PostgreSQL
06 / 07[ ACTIVE SYSTEM ]
Case Study

AI Content Pipeline

Automated content generation system running inputs through LLM chains for drafting, editing, and formatting. Human-in-the-loop review and version tracking built in.

Verified Metrics & Impact
  • -3-stage LLM chain
  • -70% reduction in drafting time
  • -Full version diff audit trail
RoleSole AI Architect
Tech StackLangChain · GPT-4 · Redis · React · FastAPI
07 / 07[ ACTIVE SYSTEM ]
Case Study

Multi-Agent Research Bot

Agentic workflow system where specialized AI agents collaborate to research topics, gather data, cross-verify facts, and produce comprehensive reports. Role-based agent teams with shared memory.

Verified Metrics & Impact
  • -4 specialized agents coordinated via LangGraph
  • -85% fact-check agreement rate
  • -Reports generated in <5min
RoleSole AI Architect
Tech StackCrewAI · LangGraph · OpenAI · Python · Redis
02 / INDEPENDENT BUILDS / AI-ASSISTED DEVELOPMENT

Ideas to Apps. One Shot.

Developed in a single one-shot Claude Code prompt · Ideas to live apps

A collection of independent projects developed in a single one-shot Claude Code prompt. From personal tools and travel utilities to food blogs and internet culture, I turn ideas into real, usable products through AI-assisted development.

Fitty screenshot
01Fitness / PWA

Fitty

A personal fitness and workout tracking PWA for planning workouts, logging sessions, tracking progress, and AI-assisted coaching.

Highlight

Workout tracking, personalized planning, progress monitoring, and AI coaching.

One-Shot Claude Code Promptfitty-weld.vercel.app
MemeSound screenshot
02Audio / Web Application

MemeSound

A meme soundboard and audio discovery website built around browsing and playing sound clips, bringing internet culture and sound effects into a simple web experience.

Highlight

Sound discovery, audio playback, and a streamlined user experience.

One-Shot Claude Code Promptmemesound.world
KhulKeKhao screenshot
03Food Blog / Content Website

KhulKeKhao

A food blog website focused on food-related content, culinary discoveries, and sharing food stories through an engaging web experience.

Highlight

Food-related content, culinary discoveries, and sharing food stories.

One-Shot Claude Code Promptkhulkekhao.world
Triply screenshot
04Travel / Group Expense Management

Triply

A custom Splitwise-style expense-sharing website built for our own travel group. Created to manage shared trip expenses, track who paid what, and simplify expense settlements among friends.

Highlight

A fun, practical solution built around the specific needs of a personal travel group.

One-Shot Claude Code Promptmh-pravas-sangh.vercel.app
Tether screenshot
05Developer Tools / Mobile Companion

Tether

A private AI coding companion that lets you control Claude Code on your MacBook from your phone. Describe a bug, let the agent work in your real local repository, review the diff and tests, and approve a push, all over an encrypted outbound tunnel with no cloud VM.

Highlight

Remote Claude Code sessions, diff review, and explicit approval before any push.

Built with Claude Codetether-gray-mu.vercel.app
03 / CAREER CHANGELOG

Career Timeline

Dec 2025 – PresentCURRENT

Datavruti

ROLE: Founding Engineer

Architecting and shipping AI-powered platforms end-to-end, including Graph (event-driven professional identity system at mygraph.id) and Estana (multi-tenant AI recruitment ATS at estana.in). Owning full-stack architecture, RAG retrieval pipelines, and production cloud infrastructure.

Impact: Founding engineer across flagship AI & identity platforms.

Jan – Dec 2025AI / ML

Silver Touch Technologies

ROLE: Senior Software Developer Engineer (AI/ML)

Architected production AI/ML systems, custom RAG retrieval engines, hybrid BM25 and vector search pipelines, LLM prompt engineering chains, and autonomous agent workflows.

Impact: Production AI systems and high-throughput retrieval pipelines.

Jan – Dec 2024FULL STACK

Silver Touch Technologies

ROLE: Senior Full Stack Developer

Engineered scalable full-stack web applications and microservices. Designed relational database schemas, REST & GraphQL APIs, and performant user interfaces.

Impact: Scalable full-stack systems and cloud infrastructure.

Jan – Dec 2023FRONTEND

Silver Touch Technologies

ROLE: React Developer

Built modern, responsive client-side web applications using React and TypeScript. Implemented modular component libraries, state management, and optimized UI performance.

Impact: Production frontend web applications and component systems.

02 / ENGINEERING OPERATING SYSTEM

How I Build

01 -

Understand the problem

Ignore surface requests. Identify real business bottlenecks, data constraints, and user workflow friction before writing a line of code.

02 -

Design the system

Architect the relational schema, vector index models, API contracts, and state flow. Build for data isolation, low latency, and zero data leakage.

03 -

Ship

Build production-ready code with strict TypeScript, clean Node/Next.js layers, containerized Docker builds, and automated cloud deployments.

04 -

Measure

Track parsing precision, TTFT (time to first token), query latency, and user pipeline conversion against real operational workloads.

05 -

Iterate

Refine RAG retrieval weights, prompt JSON schemas, and candidate re-rankers using empirical production logs.

06 -

Own the outcome

Single-point engineering accountability - taking full ownership from database migrations to user interface and uptime.

04 / SERVICES & CAPABILITIES

What I Build & Deliver

01AI Systems & Agents

AI Systems Engineering

Production RAG systems, AI agents, retrieval pipelines, and LLM-powered applications.

Production RAG · AI Agents · Retrieval Pipelines · LLM Applications
02End-to-End SaaS

Full-Stack Product Development

End-to-end SaaS engineering, APIs, databases, and complete product delivery.

SaaS Architecture · APIs & Databases · Next.js & Node.js · Product Delivery
03Scale & Reliability

Infrastructure & System Design

Scalable architectures, database design, performance optimization, and production deployments.

System Architecture · Database Design · Docker & Cloud · Performance
04Founders & Technical Leaders

Technical Product Consulting

Architecture decisions, MVP scoping, feasibility assessments, and AI integration planning.

Architecture Decisions · MVP Scoping · Feasibility · AI Integration
04 / TECHNICAL CAPABILITIES

Engineering Stack

01 / 04[ BUILD ]
4 Tech

BUILD STACK

Production web applications, server-rendered dashboards, and type-safe full-stack architectures.

Core Specializations & Competencies
  • -Modern Next.js App Router and React Server Components
  • -End-to-end type-safe API and application architecture
  • -High-performance client state management and UX
Technologies
ReactNext.jsTypeScriptNode.js
02 / 04[ AI ]
5 Tech

AI STACK

Production retrieval pipelines, multi-agent orchestration, and structured LLM applications.

Core Specializations & Competencies
  • -Hybrid vector search and dense embedding retrieval
  • -Multi-agent workflows with LangGraph and CrewAI
  • -Structured JSON schema enforcement and prompt chains
Technologies
LLMsRAGVector SearchLangGraphCrewAI
03 / 04[ SYSTEMS ]
5 Tech

SYSTEMS STACK

Multi-tenant database design, event-driven streaming, caching queues, and resilient API services.

Core Specializations & Competencies
  • -Multi-tenant schema isolation and relational data modeling
  • -Event-driven timelines and real-time streaming architectures
  • -Redis caching layers and high-concurrency API design
Technologies
PostgreSQLPrismaRedisAPIsEvent-Driven Architecture
04 / 04[ SHIP ]
4 Tech

SHIP STACK

Containerized environments, automated deployment pipelines, and reliable cloud operations.

Core Specializations & Competencies
  • -Containerized application environments with Docker
  • -Automated CI/CD pipelines with GitHub Actions
  • -Cloud infrastructure management and zero-downtime rollouts
Technologies
DockerCI/CDCloud InfrastructureProduction Deployments
05 / ABOUT & CONVICTIONS

Engineering Philosophy

I'm Lucky Patel, a Founding Engineer at datavruti, building AI-powered products and production systems from architecture through deployment.

I focus on shipping production software that solves genuine operational bottlenecks — taking full ownership of system reliability, data integrity, and end-user impact.

// ENGINEERING MANIFESTO

"Own the data. Design for real workflows. Build systems that work in production."

01

First-Party Data Moats

Rented API wrappers get commoditized quickly. True defensibility comes from domain-specific data pipelines, custom indexing, and proprietary retrieval architectures.

02

Event-Driven Identity

Careers are continuous event streams rather than static documents. Professionals deserve dynamic, verifiable career timelines backed by evidence over flat resume bullet points.

03

Full-Stack Ownership

Single-point engineering accountability — taking end-to-end responsibility from database architecture and data pipelines to interface design, deployment, and operational reliability.

04

Workflow-Scale Agents

Designing autonomous multi-agent systems that solve high-friction operational workflows and complex tasks rather than trivial chat interactions.

06 / CONTACT

Start a Conversation

Building a product, need a founding engineer, or want to integrate AI architectures into your operational stack? Let's build things that work.

Emaillbp7198@gmail.com
Phone+91 9824697672
LocationAhmedabad, Gujarat, India
Elsewhere
LinkedInGitHubInstagramTwitter