DEV Community: Rishabhh Pandey

How I Built AlumIQ A Production Ready Mern Stack Generative Ai Alumni Intelligence Platform

Rishabhh Pandey — Tue, 27 Oct 2026 15:10:38 +0000

Existing Alumni Management Tools Are Either Too Basic Just A Static Directory Or Too Complex Enterprise Erps Most Universities And Alumni Networks Still Rely On Spreadsheets To Manage Chapters Track Job Referrals And Verify Credentials There Is No Middle Ground

So I Built AlumIQ A Full Stack Alumni Intelligence Platform That Combines Real Time Community Engagement With Generative Ai And Neural Matching

Here Is The Complete Breakdown Of How It Works

━━━━━━━━━━━━━━━━━━

🛠️ The Tech Stack

⚛️ Frontend - React Js Vite Tailwind Css V4

🟢 Backend - Node Js Express Js

🍃 Database - Mongo Db Mongoose

⚡ Real Time - Socket Io

🤖 Ai Layer - Groq Llm Openai Gpt Oss 120b

🧠 Neural Matching - Tensor Flow Js Universal Sentence Encoder

☁️ Image Upload - Cloudinary Plus Multer

📧 Email And Otp - Nodemailer

🔴 Caching And Deduplication - Redis Ioredis

🔐 Authentication - Jwt Bcryptjs Otp

━━━━━━━━━━━━━━━━━━

🎯 The 10 Core Modules

1️⃣ Dashboard

Real Time Kpi Cards Showing Total Alumni Active Jobs Upcoming Events Chapters And Verified Alumni

All Data Is Fetched Live Using Mongo Db Aggregation Pipelines

2️⃣ Authentication

Jwt With Access Plus Refresh Tokens

A Secure Session Aware Axios Interceptor Auto Logs Out Users If The Token Expires

3️⃣ Alumni Management

Full Crud With Profile Details Skills And Auto Filled Geographic Coordinates

Admins Can Compare Up To 4 Alumni Side By Side Schedule Re Verifications And Perform Bulk Actions

4️⃣ Chapter Hub

Regional And Industry Based Chapters With Join Requests Member Approvals And Co Admin Assignment

Features Discussion Boards Cloudinary Gallery Uploads And A Leaflet Powered Global Heatmap Of Chapters

5️⃣ Job Board

Full Crud For Job Postings With A Redis Based Deduplication System To Track Unique Views

Prevents Users From Artificially Inflating Views

Applicants Can Shortlist And Save Jobs

6️⃣ Event Management

Create Events Physical Or Virtual Manage Rsvps With Waitlists And Check In Attendees

The Rsvp Flow Is Powered By Socket Io When A User Rsvps The Attendee Count Updates Instantly For Everyone On The Page

7️⃣ Mentorship Hub

Manage Mentor And Mentee Sessions With Feedback Ratings Action Items And Goal Tracking

8️⃣ Contribution Tracking

Track Alumni Giving Back Mentorship Hiring Donation Etc

Includes Visual Breakdowns By Type And An Ai Generated Economic Impact Narrative

9️⃣ Verification Hub

Verify Alumni Credentials Education Employment Identity Using An Immutable Audit Log

If Records Do Not Match The System Flags Them As Under Review Rather Than Auto Blocking Allowing Admins To Manually Review And Add Rejection Reasons

🔟 Directory And Directory Hub

A Central Command Center With A Watchlist Pinning

Features A 24 Hour Timeline Scrubber For Real Time Activity Replay Sensitive Zones High Value Alumni Clusters And A 30 Day Forecast Strip

━━━━━━━━━━━━━━━━━━

🤖 The Generative Ai And Neural Layer

Every Module Has A Dedicated Ai Service Powered By Groq Llm But I Did Not Just Rely On Prompt Engineering I Combined It With Tensor Flow Universal Sentence Encoder 512 Dim Embeddings

Here Are The Ai Capabilities

🎓 Alumni Ai - Bio Writer Industry Suggester Skills Suggester Matchmaker Career Path Advisor

🏛️ Chapter Ai - Description Writer Event Idea Generator Growth Advisor Recap Writer

💼 Job Ai - Jd Writer Salary Advisor Matchmaker Market Trends Overlap Detection

📅 Event Ai - Description Writer Invite Writer Schedule Advisor

🤝 Mentorship Ai - Mentor Bio Matchmaker Agenda Writer Session Summary

🛡️ Verification Ai - Regulatory Explainer Rejection Narrator Batch Grade Justification

🌟 Contribution And Directory Ai - Network Explainer Industry Storyteller Sensitivity Advisor

🧠 The Hybrid Ai Approach Tensor Flow Plus Groq

For Matchmaking Alumni Jobs Mentors I First Use Tensor Flow To Calculate The Cosine Similarity Between Text Embeddings Finding The Top Mathematically Similar Profiles

Then I Pass Those Pre Filtered Candidates To Groq Llm To Explain Why They Match And Generate Outreach Strategies

━━━━━━━━━━━━━━━━━━

💡 Key Engineering Decisions

🛡️ Json Mode - Every Ai Prompt Forces The Llm To Return Valid Json This Prevents Parsing Errors And Broken Ui

🧹 Clean Value Helper - The Ai Sometimes Returns Placeholder Values Like N A Or A Hyphen This Helper Sanitizes Them Before They Reach The Frontend

🔄 Post Processing - After Every Ai Response Arrays Are Filtered To Remove Invalid Values Generation Plus Cleanup Is The Reliable Pattern

⚡ Redis For View Deduplication - In The Jobs Module I Used Redis With A 1 Hour Ttl To Prevent Fake View Inflation If Redis Goes Down The App Gracefully Degrades

📡 Room Based Socket Io - For Events I Did Not Broadcast Rsvp Updates To Everyone Clients Join Specific Rooms This Keeps Bandwidth Low

☁️ Stateless Image Uploads - Instead Of Saving Avatars To Local Disk I Used Multer Memory Storage And Streamed The Buffer Directly To Cloudinary

━━━━━━━━━━━━━━━━━━

🔐 Security Architecture

🔑 Jwt Authentication With Access And Refresh Tokens

📧 Email Otp Verification 6 Digits 1 Min Expiry Required Before Login

🔒 Optional 2fa Via Email Otp

🛡️ Password Reset Via Otp And Short Lived Jwt Purpose Password Reset

⏱️ Rate Limiting 500 Requests Per 15 Min Global 50 Requests Per 15 Min For Auth Endpoints

🚦 Helmet Js For Http Headers Cors With Credentials

🔐 Bcrypt Password Hashing 10 Salt Rounds

📝 Immutable Audit Logs For Verifications

━━━━━━━━━━━━━━━━━━

🎨 Ui Design Philosophy

✨ Glassmorphism On Every Card Using Bg Black 40 Backdrop Blur Md Border 2 Border White 10 Rounded 3xl

🟢 Emerald Gradient Theme Across All Pages From Black Via 001a0f To Black

🔠 Uppercase Bold Typography With Tracking

🔽 Custom Dropdowns And Modals Using React Portals And Get Bounding Client Rect To Prevent Clipping

📐 Corner Brackets On Every Card That Light Up On Hover

🎬 Staggered Animations For Table Rows And Lists

📱 Fully Responsive Design

━━━━━━━━━━━━━━━━━━

💡 Key Challenges And Learnings

🚧 Challenge 1 The Auth Flow Edge Case

When Building The Middleware I Had To Ensure That An Expired Or Malformed Jwt Did Not Just Crash The App But Returned Specific Error Codes The Axios Interceptor On The Frontend Listens For These Specific Codes To Handle The Redirect Cleanly

🚧 Challenge 2 Verification Data Integrity

When An Alumni Employer Record Did Not Match I Could Not Just Auto Block Them I Designed A System Where The State Moves To Under Review

I Built An Audit Log Array Into The Mongoose Schema That Records Every Action The Admin Who Did It And A Timestamp This Provides A Trust But Verify Transparency Trail

🚧 Challenge 3 Ai Matchmaking Reliability

Pure Llms Are Bad At Math They Cannot Reliably Find The Most Similar Profile

By Shifting The Mathematical Burden To Tensor Flow Embeddings And The Reasoning Burden To Groq I Achieved A Highly Accurate And Explainable Matchmaking System

🚧 Challenge 4 Multi Role Permission

Admin And User Roles Need Different Access Levels

This Was Solved With A Protect Middleware Checking Jwt Validity Plus A Custom Role Check Middleware Ensuring Only Admins Can Access Certain Endpoints

━━━━━━━━━━━━━━━━━━

🎯 What Comes Next

I Am Currently Working On Adding Voice Based Ai Commands To AlumIQ The Goal Is To Let Admins Speak Commands Like Generate A Mentorship Agenda For John And Sarah Focusing On System Design And Have The Ai Auto Fill The Forms

Other Planned Features Include

📱 Real Time Alumni Chat Using Socket Io

📲 Mobile App Using React Native

📊 Advanced Predictive Analytics For Alumni Donations

━━━━━━━━━━━━━━━━━━

📌 Final Thoughts

Building AlumIQ Taught Me That Production Ai Is Not Just About Calling An Llm Api It Is About Prompt Engineering Error Handling And Combining Different Ai Models Like Tensor Flow Plus Groq To Get The Best Of Both Worlds

The Biggest Lesson Was That Reliability Matters More Than Cleverness A Simple Json Mode Plus Cleanup Logic Beats An Elaborate Prompt That Sometimes Fails

The Second Lesson Was That Real Time Features Need Careful Architecture Broadcasting To All Clients Kills Performance Room Based Subscriptions Are The Way

━━━━━━━━━━━━━━━━━━

🔗 Interested In The Full Source Code Or Getting This Project For Your Portfolio

🌐 Check Out The Project Stack Website

🌐 https://www.theprojectstack.com/

Happy To Answer Any Questions About The Architecture Or Ai Integration Drop A Comment Below

How I Built AdCampaignManager A Production-Ready MERN Stack Generative AI Advertising SaaS Platform

Rishabhh Pandey — Wed, 14 Oct 2026 06:03:28 +0000

Existing advertising campaign tools are either too expensive or too complex. Most small agencies and freelancers still manage campaigns in spreadsheets. There is no middle ground.

So I Built AdCampaignManager — A Full-Stack Advertising Campaign Management Platform That Combines Real-Time Data With Generative AI.

Here Is The Complete Breakdown Of How It Works.

━━━━━━━━━━━━━━━━━━

🛠️ The Tech Stack

Frontend - React.Js Plus Vite Plus Tailwind CSS v4
Backend - Node.Js Plus Express.Js
Database - MongoDB Plus Mongoose
Real-Time - Socket.IO
AI Layer - Groq LLM openai/gpt-oss-120b
Image Upload - Cloudinary Plus Multer
Email - Nodemailer
Scheduled Tasks - Node-Cron
Authentication - JWT Plus Bcrypt Plus OTP

━━━━━━━━━━━━━━━━━━

🎯 The 10 Core Modules

Dashboard

Real-Time KPI Cards Showing Impressions Clicks Conversions Revenue CTR CPC ROI And CVR. All Data Fetched Live From MongoDB Aggregation Pipelines.

Authentication

Register Login Email OTP Verification 2FA Password Reset Session Management. JWT With Access Plus Refresh Tokens Stored In localStorage.

Campaign Management

Full CRUD With Status Lifecycle Draft To Scheduled To Running To Paused To Completed To Cancelled. Duplicate Compare Side-By-Side Creative Upload Via Cloudinary.

Audience Management

Age Range Gender 50 Countries 50 Cities 50 Interests 50 Languages And Device Targeting. Overlap Detection Between Segments.

Analytics And Insights

Real-Time KPIs Engagement Heatmap Hour By Day Traffic Sources Funnel Analysis Campaign Leaderboard.

Subscription Plans

12 Plus Premium Tiers Subscribe Auto-Renew Corporate Plans Usage Analytics.

Reviews And Moderation

5-Star Ratings Sentiment Badges Photo Uploads Admin Approval Workflow.

Team Collaboration

Real-Time Chat Via Socket.IO Kanban Boards Member Roles Activity Feed.

Support Tickets

Categories Priorities Reply Threads Internal Notes Knowledge Base.

AI Insights Dashboard

Platform-Wide KPIs Anomaly Alerts AI Narratives.

━━━━━━━━━━━━━━━━━━

🤖 The Generative AI Layer

Every Module Has Its Own Dedicated AI Service Powered By Groq LLM openai/gpt-oss-120b.

Here Are The 10 AI Services

Analytics AI

Narrate Insights Detect Anomalies Forecast Revenue For 30 60 90 Days Generate Report Advise KPI Explain Trend Reallocate Budget Analyze Creative.

Campaign AI

Write Copy Suggest Goal Optimize Budget Generate Name Quality Check Suggest Schedule.

Audience AI

Recommend Audience Suggest Interests Estimate Reach Optimize Demographics Build Lookalike Resolve Overlap.

Plan AI

Recommend Plan Write Benefits Advise Upgrade Detect Renewal Risk Suggest Pricing Build Corporate Plan.

Review AI

Sentiment Analysis Fake Review Detection Summarize Reviews Draft Admin Reply Moderate Photo Extract Highlights.

Insights AI

Narrate Platform Detect Anomalies Forecast Revenue Generate Report Advise KPI Explain Trend.

Support AI

Classify Ticket Draft Reply Detect Duplicate Suggest FAQ Predict Resolution Time Auto-Write FAQ.

Team AI

Summarize Chat Auto-Assign Tasks Find Meeting Time Productivity Insights Suggest File Tags Recommend Lead.

━━━━━━━━━━━━━━━━━━

💡 Key Engineering Decisions

JSON Mode

Every AI Prompt Forces The LLM To Return Valid JSON. This Prevents Parsing Errors And Broken UI. Without JSON Mode The AI Sometimes Returns Markdown Which Crashes The Frontend Parser.

cleanValue Helper

The AI Sometimes Returns Placeholder Values Like N/A Or A Hyphen Or Empty Strings. This Helper Sanitizes Them Before They Reach The Frontend.

Post-Processing

After Every AI Response Arrays Are Filtered To Remove Invalid Values. Generation Plus Cleanup Is The Reliable Pattern.

Real-Time Sync

Socket.IO Room-Based Subscriptions For Team Chat. Each Client Joins Only The Rooms It Needs Using team Colon teamId Format. This Keeps Bandwidth Low And Updates Instant.

Scheduled Automation

Node-Cron Runs Two Jobs. Every Minute It Simulates Running Campaigns To Generate Realistic Impressions Clicks Conversions Spend And Revenue. Daily At Midnight It Aggregates Previous Day Analytics Into A Single Snapshot Document For Fast 30 60 90 Day Queries.

━━━━━━━━━━━━━━━━━━

🔐 Security Architecture

JWT Authentication With Access And Refresh Tokens
Email OTP Verification 6 Digits 1 Min Expiry
Optional 2FA Via Email
Password Reset Via OTP And Short-Lived JWT
Rate Limiting 500 Requests Per 15 Min Global 50 Requests Per 15 Min Auth
Helmet.Js Security Headers
CORS With Credentials
Cloudinary File Type Validation
Bcrypt Password Hashing 10 Salt Rounds
Environment Variables For All Secrets

━━━━━━━━━━━━━━━━━━

🎨 UI Design Philosophy

Glassmorphism On Every Card Using backdrop-blur-xl border-2 border-white/10 rounded-3xl
Green Gradient Theme from-black via-[#021d0f] to-black Across All Pages
Uppercase Bold Typography With tracking-[0.2em]
Custom Dropdowns With Animated Chevrons
Modal Portals With Backdrop Blur
Toast Notifications With Color-Coded Types
Corner Brackets On Every Card
Floating Icon Animations
Staggered Fade-In For Table Rows
Fully Responsive Design

━━━━━━━━━━━━━━━━━━

💡 Key Challenges And Learnings

Challenge 1 — AI Output Reliability

The LLM Sometimes Returns Invalid JSON Or Placeholder Values. This Was Fixed With A Three-Layer Approach. First Force JSON Mode At The API Level. Second Sanitize Every Value With The cleanValue Helper. Third Filter Arrays During Post-Processing.

Challenge 2 — Real-Time Team Chat

The Team Members Needed Live Chat Without Refresh. This Was Solved With Socket.IO Room-Based Subscriptions. Each Client Joins Only The Rooms It Needs.

Challenge 3 — Fast Analytics Queries

The Analytics Collection Grew Huge. This Was Solved With Daily Snapshot Cron Job That Aggregates Data Into A Lightweight Collection For Fast Queries.

Challenge 4 — Multi-Role Permission

Admin And User Roles Need Different Access Levels. This Was Solved With owner Field Check Plus isAdmin Boolean Plus Backend authorize Middleware.

Challenge 5 — Campaign Lifecycle

Campaigns Need To Move Through Multiple States. This Was Solved With Auto Transitions From Draft To Scheduled To Running To Paused To Completed To Cancelled.

Challenge 6 — Image Uploads In Serverless

Local Disk Storage Breaks In Serverless Deployments. This Was Solved With Cloudinary Streaming API. The File Buffer Goes Directly To CDN.

━━━━━━━━━━━━━━━━━━

🎯 What Comes Next

I Am Currently Working On Adding Voice-Based AI Assistant To AdCampaignManager. The Plan Is To Let Users Speak Commands Like Create A New Campaign For Sneakers With 5000 Rupees Daily Budget And The AI Will Auto-Fill Everything.

Other Planned Features Include

Real-Time Collaboration With Cursor Tracking
Mobile App Using React Native
Public API For Third-Party Developers
Advanced Predictive Analytics Dashboard

━━━━━━━━━━━━━━━━━━

📌 Final Thoughts

Building AdCampaignManager Taught Me That Production AI Is Not Just About Calling An LLM. It Is About Prompt Engineering Error Handling Post-Processing And Delivering Clean Data To Users.

The Biggest Lesson Was That Reliability Matters More Than Cleverness. A Simple JSON Mode Plus Cleanup Logic Beats An Elaborate Prompt That Sometimes Fails.

The Second Lesson Was That Real-Time Features Need Careful Architecture. Broadcasting To All Clients Kills Performance. Room-Based Subscriptions Are The Way.

━━━━━━━━━━━━━━━━━━

Interested In The Full Source Code Or Getting This Project For Your Portfolio?

🌐 https://www.theprojectstack.com/

Happy To Answer Any Questions About The Architecture Or AI Integration. Drop A Comment Below!

How I Built AirSense - A Production-Ready MERN + Generative AI Air Quality Ecosystem

Rishabhh Pandey — Mon, 12 Oct 2026 12:23:20 +0000

Existing Air Quality Dashboards Just Show Numbers. No Context. No Forecasts. No Actionable Insights.

So I Built AirSense — A Full-Stack Air Quality Monitoring Ecosystem That Combines Real-Time Data With Generative AI.

Here Is The Complete Breakdown Of How It Works.

🛠️ The Tech Stack

Frontend - React.Js + Tailwind CSS
Backend - Node.Js + Express.Js
Database - MongoDB + Mongoose
Real-Time - Socket.IO
AI Layer - Groq LLM
Cloud - Cloudinary + WAQI API
Authentication - JWT With 2FA And OTP Verification

🌬️ Core Features

Real-Time Monitoring

A Multi-Station AQI Dashboard With Live Updates Via Socket.IO. Color-Coded Heatmaps Across All Monitoring Stations. A 24-Hour Timeline Scrubber And Trend History. Every New Reading Pushes Instantly To Connected Clients Without A Page Refresh.

Generative AI Layer Powered By Groq LLM

AI Insight Narrator — Plain-English AQI Explanations With Exact Numbers
Anomaly Explainer — Detects Unusual Pollution Patterns And Suggests Causes
Health Advisory Writer — Real-Time Precautions For Sensitive Zones Like Schools And Hospitals
Pollution Source Narrator — Explains Why A Specific Pollutant Dominates At A Station
Compliance Summarizer — Regulatory Insights For WHO US EPA CPCB And EU Standards
Report Writer — AI-Generated Executive Reports For Any Time Period
KPI Advisor — Recommends Which Metrics To Focus On Based On Live Data
Trend Explainer — Breaks Down Why A Metric Moved Up Down Or Stayed Stable

Forecasting Engine

Statistical Time-Series Forecasting Using Linear Regression. Predictions For 12-Hour 7-Day And 30-Day Horizons. Every Forecast Is Enhanced With AI-Generated Explanations And Confidence Scoring. Also Supports Side-By-Side Forecast Comparison Across Multiple Stations.

Multi-Standard Compliance

Instantly Check Pollutant Levels Against WHO US EPA CPCB And EU Standards. Each Reading Is Graded As Pass Or Fail With Exceedance Percentage Calculated Automatically.

Pollution Source Attribution

AI Analyzes And Attributes Pollution To Sources Like Traffic Industrial Construction Crop Burning Natural And Other. Each Source Comes With Impact Metrics For Population Health Economy And Carbon Footprint.

🤖 The AI Layer In Detail

Every AI Feature Runs On Groq LLM Using Two Models. The Primary Model Is openai/gpt-oss-120b For Complex Reasoning And The Secondary Model Is llama-3.1-8b-instant For Faster Simpler Tasks.

Key Engineering Decisions

JSON Mode

Every Prompt Forces The LLM To Return Valid JSON. This Prevents Parsing Errors And Broken UI. Without JSON Mode The AI Sometimes Returns Markdown Or Plain Text Which Crashes The Frontend Parser.

cleanValue Helper

The AI Sometimes Returns Placeholder Values Like A Hyphen Or N/A Or Empty Strings. This Helper Sanitizes Them Before They Reach The Frontend. It Also Filters Out The String Null And Unknown Values.

sessionSeed

A Random 6-Character String Is Injected Into Every Prompt. This Prevents The AI From Repeating The Same Answer On Every Call. Without This Trick The Model Would Return Identical Responses Due To Caching And Temperature Behaviour.

Post-Processing

After Every AI Response Arrays Are Filtered To Remove Invalid Values. Generation Plus Cleanup Is The Reliable Pattern. Every Array Field Like Insights Causes Fixes Recommendations Factors And Actions Goes Through This Cleanup Step.

Temperature Tuning

Different AI Services Use Different Temperature Settings. Creative Tasks Like Copywriting Use Higher Temperature Around 0.85 While Deterministic Tasks Like Station Description Writing Use Lower Temperature Around 0.3.

📊 Forecasting Is Not AI But Enhanced By It

The Prediction Itself Uses Linear Regression Which Is A Statistical Method. Not A Neural Network. Not A Transformer. Just Classic Maths.

But Every Forecast Is Enhanced With -

AI-Generated Explanations Of The Trend
Confidence Scoring That Decreases With Horizon
Health Advisory Recommendations Based On Predicted AQI
Weather Impact Analysis

This Hybrid Approach Is What Modern Production Apps Actually Use. Not Everything Needs To Be AI. Sometimes A Simple Statistical Model Plus AI Explanation Is Better Than An Overcomplicated Neural Network.

🔐 Security Architecture

JWT Authentication With 2FA
OTP Verification Via Email For Registration And Password Reset
Role-Based Access Control For Admin And User Roles
Session Management With Full Audit Trail
Password Hashing With Bcrypt
Rate Limiting On Sensitive Endpoints
CORS Configuration For Frontend Only
Environment Variables For All Secrets

💡 Key Challenges And Learnings

Challenge 1 - AI Output Reliability

Sometimes The AI Returned Placeholder Values Empty Arrays Or Even Broke JSON Format. This Was Fixed With A Three-Layer Approach. First Force JSON Mode At The API Level. Second Sanitize Every Value With The cleanValue Helper. Third Filter Arrays During Post-Processing.

Challenge 2 - Real-Time Sync

The Dashboard Needed Live Updates Without Refreshing. This Was Solved With Socket.IO Room-Based Subscriptions Using station-id Format. Each Client Joins Only The Rooms It Needs. This Keeps Bandwidth Low And Updates Instant.

Challenge 3 - Stateless Uploads

Storing Images On The Server Would Break In Serverless Deployments. This Was Solved By Using Cloudinary Streaming API Instead Of Local Disk Storage. The File Buffer Goes Directly To Cloudinary And Only The HTTPS URL Is Stored In MongoDB.

Challenge 4 - Multi-Standard Compliance Logic

Each Regulatory Body Has Different Limits For Each Pollutant. This Was Solved With A Central LIMITS Object That Maps Every Standard-Pollutant Combination To Its Threshold Value. Adding A New Standard Only Requires Updating This One Object.

🎯 What Comes Next

I Am Currently Working On Adding LSTM Neural Networks For Forecasting To Push Accuracy Higher Than Linear Regression. The Plan Is To Train The Model On Historical AQI Data And Use TensorFlow.Js For Direct Node.Js Integration Without Needing A Separate Python Service.

Other Planned Features Include -

Real Payment Gateway Integration
Mobile App Using React Native
Public API For Third-Party Developers
Advanced Historical Analytics Dashboard

📌 Final Thoughts

Building AirSense Taught Me That Production AI Is Not Just About Calling An LLM. It Is About Prompt Engineering Error Handling Post-Processing And Delivering Clean Data To Users.

The Biggest Lesson Was That Reliability Matters More Than Cleverness. A Simple JSON Mode Plus Cleanup Logic Beats An Elaborate Prompt That Sometimes Fails.

Interested In The Full Source Code Or Getting This For Your Portfolio?

🌐 https://www.theprojectstack.com/

Happy To Answer Any Questions About The Architecture Or AI Integration. Drop A Comment Below!