Saad Rasheed logo
Saad Rasheed
Operations Automation

Stop Losing Tasks in
Messy Email Threads

Turn internal email comments into fully categorized, prioritized, and assigned Asana tasks instantly.

"Tired of client requests falling through the cracks because your team is overwhelmed by their inbox?"

FrontAppn8nAsanaClaude AI
The Problem
🧊

Leads go cold

Tasks created too late because no one saw the email comment in time. Follow-ups were missed entirely.

📋

Manual overhead

Team members copy-pasted task details from FrontApp into Asana by hand — slow, error-prone, and draining.

🔀

No prioritization

Urgent items sat in the same queue as routine tasks with no way to automatically flag what needed to go out today.

How It Works
01

Comment Added

Team member leaves an internal comment in a FrontApp shared email thread.

02

Parsed & Classified

n8n webhook fires; Claude AI reads the emoji prefix and extracts the task details.

03

Member Looked Up

The @mention is matched to an Asana email and GID via a lookup table in n8n.

04

Task Created

Asana task created in the right list — To Do or Urgent — and assigned instantly.

Tech Stack
FrontApp

Communication Layer

Monitors internal comments on emails and shared workspace discussions. Serves as the human-facing trigger point for every task.

  • Internal comments
  • Email threads
  • Shared discussions
n8n

Automation Backbone

Orchestrates the entire workflow — receives the webhook, calls Claude AI, resolves the user, and creates the Asana task.

  • Webhook triggers
  • Claude AI parsing
  • Logic & routing
Asana

Task Management Layer

Receives structured task data and creates cards in the correct list with proper assignments and due-date flags.

  • To Do list
  • Urgent — due today
  • Auto-assignment
Live Example
■ @bruce follow up with the client regarding the invoice.
✓ Task CreatedTo Do — TasksAssigned: Bruce
→
"Follow up with the client regarding the invoice"
To Do — Tasks
■■ @bruce This needs to go out today — urgent!
⚡ Urgent TaskDue TodayAssigned: Bruce
→
"This needs to go out today"
Urgent — Due Today
Key Features
■

Emoji Syntax

Simple ■ and ■■ prefixes let anyone trigger tasks without training or a new tool.

⚡

Instant Routing

Urgent vs. standard tasks automatically land in the correct Asana list — no manual sorting.

@

Mention Mapping

@ mentions resolve to real Asana users via a lookup table maintained in n8n.

🤖

AI-Powered Parsing

Claude AI extracts clean task descriptions from messy, natural-language comments.

✓

Auto-Assignment

Every task is assigned to the right person with zero manual steps.

🔒

Scope Control

Only triggers on shared workspace discussions — private threads are excluded for safety.

Ready to scale?

Built for a high-volume operations team. Custom emoji syntax, any project management tool, live in days.

  • Custom emoji syntax & routing
  • Any project management tool
  • Live in days, not months
Lead Conversion

A Voice Agent That
Never Misses a Call

Answers inbound calls, qualifies leads with custom questions, and books confirmed meetings onto your calendar without any human input.

"How many high-value clients are you losing simply because no one picked up the phone after hours?"

VAPIn8nCalendly
The Problem
🧊

Leads go cold

No one picks up after hours or on weekends. Inbound interest dies before it reaches the sales team.

🔁

Repetitive calls

Sales reps spent hours repeating the same qualifying questions for every new lead — high cost, low leverage.

📅

Missed bookings

Manual scheduling introduced delays and drop-offs between the first conversation and a confirmed meeting.

How It Works
01

Call Received

Customer calls in; the AI agent picks up immediately — any time, any day.

02

Lead Qualified

Agent asks targeted questions to understand the customer's need before booking.

03

Slot Selected

n8n checks live Calendly availability and offers real time-slots to the caller.

04

Meeting Booked

Booking confirmed; calendar invite and meeting notes sent automatically.

Tech Stack
VAPI

Voice AI Engine

Powers the real-time voice conversation. Handles speech recognition, natural language understanding, and tool execution mid-call.

  • Real-time voice calls
  • Tool calling
  • Natural conversation
n8n

Automation Backbone

Orchestrates the workflow. Receives triggers from VAPI, checks Calendly availability, and schedules confirmed meetings.

  • Webhook triggers
  • API integrations
  • Logic & routing
Calendly

Scheduling Layer

Provides live availability slots and creates confirmed bookings with automatic calendar invites for both parties.

  • Live availability
  • Auto booking
  • Calendar invites
Live Demo
AI ReceptionistLIVE
Agent
Hi! Good morning. What brings you here today?
User
I want to build an AI-powered app for my business.
Agent
That's great! May I have your full name to personalize the booking?
User
Sure, I'm Alex.
Agent
And your email address?
User
[email protected]
Agent
✅ Meeting booked for Tuesday at 2:00 PM Eastern Time. Check your email for the invite!

What just happened?

Greeted & understood the need
Asked qualifying questions
Collected name & email
Fetched live calendar slots
Booked meeting via Calendly
Sent calendar invite automatically
Generated & attached meeting notes
Key Features
🎙️

Natural Voice

Sounds human, responds with warmth and confidence — callers can't tell it's automated.

🧠

Smart Qualifying

Asks the right questions before booking so every meeting comes with context.

📋

Strict Booking Flow

Collects name and email before checking availability — no incomplete bookings.

📅

Live Availability

Checks real calendar slots in real time via n8n — no double bookings, no guessing.

📝

Auto Meeting Notes

Summarizes the call and attaches notes directly to the Calendly booking.

⚡

Zero Human Input

Fully automated from first word to confirmed invite. No one needs to be available.

Your business deserves an AI that never sleeps.

Built for a leading agency. Custom voice, any industry, any language — live in weeks.

  • Custom voice & personality
  • Any industry, any language
  • Live in weeks, not months
Customer Database Builder

Log Every Caller & Build
Your Marketing Database

A 24/7 AI phone system that answers inbound calls, logs every customer's details directly into your CRM, and creates a database for future marketing campaigns.

"What if every single person who called your store was automatically saved to a database, ready for your next marketing campaign?"

VAPIn8nVtiger CRMTelegram
The Problem
📵

Zero Caller Capture

Dozens of customers called daily, but their details were never captured. Every hang-up was a lost marketing opportunity.

🗂️

No CRM data

Every inbound enquiry was handled verbally with nothing recorded. The team had no history of who called, what they needed, or when.

💬

Manual follow-up

SMS follow-ups were done by hand. There was no central place to see calls, manage customer messages, or track who had been contacted.

How It Works
01

Call Received

AI agent answers every inbound call — whether the store is open or closed.

02

Data Collected

Agent gathers customer name, query, and consent for SMS follow-up.

03

CRM + SMS + Log

Customer is written to Vtiger CRM, SMS sent if consented, call logged to Telegram.

04

Database Built

Caller details are saved into a central database for future marketing campaigns and outreach.

Tech Stack
VAPI

Voice AI Engine

Powers the real-time phone conversation. Handles natural speech, collects structured data from callers, and executes business logic mid-call.

  • Inbound call handling
  • Natural conversation
  • Hours-aware behaviour
n8n

Automation Backbone

Orchestrates everything after the call — CRM writes, SMS dispatch, Telegram logging, and call forwarding logic based on store hours.

  • Vtiger CRM writes
  • SMS trigger
  • Telegram log dispatch
Vtiger + Telegram

CRM & Comms Layer

Vtiger stores every customer record. Telegram serves as the team's unified inbox — call summaries arrive here, and two-way SMS replies are managed from the same thread.

  • Full customer records
  • Call summaries logged
  • Two-way SMS from Telegram
Live Example
AI Phone AgentAFTER HOURS
Agent
Thank you for calling. Our store is currently closed. May I take your name and what you're looking for today?
User
I'm James, I need a RAM upgrade for my laptop.
Agent
Got it James. Can I get your phone number so we can follow up when we open?
User
Sure — 786-555-0192. And yes, you can text me.
Agent
Perfect, James. I've logged your enquiry and you'll get a text from us. We open tomorrow at 9 AM — speak soon!

What just happened?

AI agent answered the call
Collected name, query & number
Got SMS consent from customer
Record written to Vtiger CRM
Follow-up SMS sent automatically
Call summary logged to Telegram
Team can reply via Telegram thread
Key Features
🕐

24/7 Lead Capture

Never miss a caller. The AI agent handles calls around the clock and consistently logs their information.

🗂️

Auto CRM Logging

Every caller's name, query, and contact details are written directly to Vtiger CRM — no data entry needed.

💬

Consent-Based SMS

If the customer agrees to receive a text, an automated follow-up SMS is sent immediately after the call ends.

📲

Two-Way SMS via Telegram

All call summaries land in a Telegram channel. The team can reply to any customer's SMS directly from that same Telegram thread.

📋

Full Call Logging

Every call — open or closed, answered or missed — is summarized and posted to Telegram for the team's records.

🔁

Zero Manual Follow-Up

From call answered to CRM record to SMS sent — the entire post-call workflow runs without anyone on the team lifting a finger.

Never miss a customer
call again.

Works for any retail store, service business, or clinic. Custom hours logic, any CRM, live in days.

  • Custom hours & routing logic
  • Any CRM integration
  • Live in days, not months
Business Intelligence

Talk to Your Database
in Plain English

Ask questions on Telegram or Slack and get formatted answers, charts, and downloadable Excel reports instantly. No SQL required.

"Are you tired of waiting days for a developer just to run a simple data report?"

SupabasePostgreSQLLLMTelegram
The Problem
🧱

SQL barrier

Non-technical team members couldn't query their own data without involving a developer — every simple question caused a delay.

⏳

Slow turnaround

Waiting for a developer to run a report meant decisions were made on stale data or gut feel rather than real numbers.

📊

No self-serve reporting

There was no way for the team to pull formatted reports or charts without custom dashboard work — and those took weeks to build.

How It Works
01

User Asks

Team member sends a plain-English question on Telegram, email, or the web interface.

02

LLM Converts

Multiple LLM calls convert the question into a safe, executable SQL query.

03

Query Runs

SQL executes against Supabase. On error, it's retried automatically with error context.

04

Answer Returned

Formatted answer with data visualizations and downloadable Excel report sent back.

Tech Stack
Supabase

Database Layer

Hosts the PostgreSQL database. Provides a secure API for querying live business data on demand with user-level access control.

  • PostgreSQL database
  • Row-level security
  • Real-time data
LLM Pipeline

NL → SQL Engine

Multiple LLM calls handle intent extraction, schema matching, SQL generation, and error recovery — producing reliable queries from free-form questions.

  • Multi-step LLM calls
  • Error retry logic
  • Schema-aware generation
Telegram / Email / Web

Multi-Platform Interface

The same pipeline runs across Telegram, email, and a web interface. Users pick whichever channel fits their workflow — the backend is shared.

  • Telegram bot
  • Email interface
  • Web chatbot
Live Example
Data BotONLINE
User
Which products had the most returns last month?
Bot
Got it — pulling from the orders table…
Bot
✅ Top 3 by returns: 1. USB-C Hub (42) 2. Laptop Stand (31) 3. Webcam Pro (27). Full report attached as Excel.
User
Why did the USB-C Hub have so many?
Bot
Checking return reason codes… Most cited: "incompatible with MacBook" (29 of 42).

What just happened?

Natural language question received
LLM converted to SQL query
Executed against Supabase
Formatted answer returned
Excel report generated & attached
Follow-up question handled in context
Key Features
💬

Plain English Queries

Anyone on the team can ask data questions in natural language — no SQL knowledge required.

📊

Excel Report Output

Every successful query generates a downloadable, formatted Excel report alongside the text answer.

🔁

Auto Error Recovery

On a SQL error, the query is retried with error context. If still unresolved, the user is prompted to clarify.

📱

Multi-Platform

Works on Telegram, email, and the web — same intelligence, whichever channel the team prefers.

🗄️

Stored Query History

Every query and its result is stored, creating a searchable history of what was asked and when.

🔒

Secure by Design

Queries run through Supabase row-level security — users only see data they're permitted to access.

Your team deserves
self-serve data.

Works with any PostgreSQL or Supabase database. Custom to your schema, live in days.

  • Any SQL database
  • Custom to your schema
  • Live in days, not months
Content Automation

Scale Your Content Without
Scaling Your Team

Enter a topic in Google Sheets and instantly generate a full video script, aligned voiceover, and YouTube description in your Drive.

"How much time are you wasting formatting scripts instead of actually recording content?"

Google SheetsApps ScriptLLMGoogle Drive
The Problem
⏳

Slow pre-production

Content creators spent hours writing scripts, voiceovers, and descriptions manually for every video — before a single frame was recorded.

🔁

Repetitive formatting

The same structure was written from scratch each time. There was no template, no consistency, and no way to speed it up without hiring writers.

📂

Disorganized outputs

Drafts lived in random docs, were hard to find, and never aligned — script timing didn't match the voiceover, and thumbnails were an afterthought.

How It Works
01

Topic Entered

Creator fills in video topic, target keywords, and key information in Google Sheets.

02

Button Clicked

"Create Documents" is clicked — Apps Script triggers the LLM pipeline immediately.

03

Content Generated

LLM generates all three documents in parallel — script, voiceover, and thumbnail brief.

04

Docs Saved

All three files are saved directly to Google Drive — organized and ready to use.

Tech Stack
Google Sheets

Input Interface

The creator's familiar workspace. Input fields for topic, keywords, and notes sit alongside a single button that kicks off the entire pipeline.

  • Topic & keyword input
  • One-click trigger
  • No new tools to learn
Apps Script + LLM

Generation Engine

Google Apps Script handles the trigger and file creation. An LLM generates each document type with its own prompt tailored to the output format.

  • Parallel document generation
  • Format-specific prompts
  • Timestamp alignment
Google Drive

Output Layer

All generated documents are saved directly to the creator's Drive — named, organized by video topic, and immediately shareable with the production team.

  • Auto-organized folders
  • Named by topic
  • Instantly shareable
Live Example
Topic: "5 AI Tools That Save 10 Hours a Week"
Keywords: AI productivity, automation tools, 2025
Key info: Focus on free tools, beginner-friendly
▶ Create Documents clicked
→
✓ Video Script — visual scenes with timestamps
✓ Voiceover Doc — aligned spoken text per scene
✓ Thumbnail & Description — brief + YouTube copy
Saved to Drive
Documents Generated
🎬

Video Script Document

Scene-by-scene breakdown with visual descriptions, camera directions, and timestamps aligned to the full video length.

🎙️

Voiceover Document

Spoken text aligned to each scene's timeframe — ready to record directly or send to a voice actor.

🖼️

Thumbnail Brief

Design direction for the thumbnail — headline, visual concept, and color guidance — alongside the full YouTube description and tags.

⚡

One-Click Trigger

No new tools, no new tabs. The entire pipeline runs from a button inside the Google Sheet the creator already uses.

📁

Auto-Organized Drive

All three documents land in a named folder in Google Drive — no manual saving, renaming, or filing.

🔄

Reusable for Any Topic

Same system, any video topic. The LLM adapts the structure and tone to the subject matter automatically.

Ship content
faster than ever.

Works for any content creator, YouTube channel, or agency. Custom to your brand voice, live in days.

  • Custom brand voice & format
  • Any content type or niche
  • Live in days, not months
Legal Risk Mitigation

Detect Contract Risks in
Seconds, Not Days

Upload any NDA to find missing clauses, ambiguous language, and risky terms — complete with suggested rewrites inserted directly as Word comments.

"Are you signing agreements blindly because professional legal review takes too long and costs too much?"

Google FormsApps ScriptLLMWord API
The Problem
⚖️

No legal resource

Businesses were signing NDAs without a lawyer reviewing them — missing clauses that left them exposed or locked into unfair terms.

⏳

Slow turnaround

When legal review did happen, it took days to get feedback. Deals stalled and momentum was lost waiting for a document to come back.

🔍

Missed gaps

Non-lawyers couldn't spot what was missing — vague termination clauses, absent jurisdiction terms, or undefined confidentiality scope went unnoticed.

How It Works
01

NDA Uploaded

User submits their NDA through a simple Google Form — no account needed.

02

Text Extracted

Apps Script parses the document and passes the full content to the LLM pipeline.

03

Clauses Analyzed

LLM identifies gaps, ambiguities, and risky language — with specific suggested fixes.

04

Word Doc Returned

Revised document delivered with highlights and all suggestions inserted as tracked Word comments.

Tech Stack
Google Forms

Submission Interface

A clean upload form — no login, no complexity. The user attaches their NDA and submits. Apps Script handles everything from there.

  • Simple upload flow
  • No account required
  • Instant processing trigger
Apps Script + LLM

Analysis Engine

Apps Script extracts the document content and runs it through the LLM. The model identifies clause issues and generates specific, actionable suggestions for each one.

  • Full NDA text extraction
  • Clause-level analysis
  • Insert / replace / delete suggestions
Word API

Output Layer

The improved document is built as a proper Word file — with highlighted text, inline comments, and all suggested edits in a format any lawyer or business owner can work with directly.

  • Tracked Word comments
  • Highlighted problem areas
  • Ready to review & sign
Live Example
📄 mutual_nda_acme_corp.docx uploaded via Google Form
Processing…
→
⚠ Missing: Jurisdiction clause (insert suggested)
⚠ Vague: "reasonable efforts" — recommend defined standard
✓ Termination clause: acceptable
⚠ Confidentiality scope: too broad — suggest narrowing
Word doc + comments returned
Key Features
🔍

Clause Gap Detection

The LLM checks for missing clauses — jurisdiction, termination, scope, remedies — and flags each one specifically.

📝

Tracked Word Comments

Every suggestion is inserted as a Word comment on the relevant text — ready to accept, reject, or discuss with a lawyer.

✏️

Specific Rewrites

For vague or risky language, the system suggests exact replacement text — not just a flag, but a fix.

⚡

Seconds, Not Days

Upload to revised Word doc in under a minute — no waiting for legal review to start the conversation.

🎯

Insert / Replace / Delete

Each suggestion is categorized — what to add, what to rewrite, and what to remove — making review straightforward.

📤

No Login Required

Upload via Google Form, receive the output by email. No account, no dashboard, no friction.

Review NDAs in
seconds, not days.

Works for any contract type. Custom clause library for your industry, live in days.

  • Custom clause library
  • Any contract type
  • Live in days, not months
Revenue Recovery

Recover Visitors Who
Leave Without Booking

Capture visitor emails before they bounce, automatically detect if they didn't complete a booking, and trigger a personalized follow-up sequence.

"How many potential clients browse your portfolio, love your work, but leave the site without ever saying hello?"

ZapierSystem.ioCalendly
The Problem
👋

Visitors leaving silently

Interested couples were browsing the photography website and leaving without any contact — no name, no email, no way to follow up.

📉

Drop-offs between email and booking

Some visitors gave their email but never made it to Calendly. There was no visibility into this gap and no way to recover those leads.

📧

Manual follow-up

When follow-up did happen, it was done by hand — inconsistent, slow, and dependent on someone remembering to send the email.

How It Works
01

Email Captured

A website pop-up collects the visitor's email before they are redirected to Calendly.

02

Booking Checked

Zapier checks whether a booking was made on Calendly after the email was submitted.

03

Gap Detected

If email was given but no booking followed, the lead is flagged automatically.

04

Sequence Triggered

System.io sends a targeted follow-up email campaign to recover the unbooked lead.

Tech Stack
Website Pop-Up

Lead Capture Layer

An email capture pop-up appears on key pages of the photography website — collecting contact details before the visitor is sent to Calendly.

  • Timed trigger
  • Exit intent option
  • Email capture before Calendly
Zapier

Automation Backbone

Connects the pop-up, Calendly, and System.io. Checks booking status after each email capture and routes accordingly — booked leads are tagged, unbooked leads enter the recovery flow.

  • Booking status check
  • Conditional routing
  • Lead tagging
System.io

Email Marketing Layer

Runs the follow-up sequence for unbooked leads — a series of warm, personalized emails designed to bring them back and complete the booking.

  • Automated sequences
  • Personalized follow-ups
  • Booking link re-engagement
Live Example
✉ [email protected] submitted pop-up form
⏳ Waited 30 mins — no Calendly booking detected
Lead flagged as unbooked
→
Day 0: "Still thinking? Here's what couples say…"
Day 2: "Only 3 dates left for your season"
Day 5: "Book a free 15-min call — no commitment"
Sequence triggered automatically
Key Features
📬

Pre-Calendly Email Capture

Email is collected before the visitor reaches Calendly — ensuring no lead is lost even if they don't book.

🔍

Booking Gap Detection

The system automatically identifies who submitted their email but didn't complete a booking — and acts on it.

📩

Automated Follow-Up Sequence

A warm, personalized multi-email sequence is sent to unbooked leads — no manual sending required.

🏷️

Lead Tagging

Every lead is tagged as booked or unbooked in the system — giving the photographer full visibility of their pipeline.

⚡

Zero Manual Effort

Once live, the entire capture-check-follow-up loop runs automatically — no one needs to monitor or send anything.

📈

Recovers Lost Revenue

Converts visitors who showed intent but didn't commit — turning a previously invisible gap into booked sessions.

Stop losing leads
that were already interested.

Works for any service business with a booking flow. Custom sequences, live in days.

  • Custom follow-up sequences
  • Any booking platform
  • Live in days, not months
E-Commerce AI

Studio-Quality Photos
Without the Photoshoot

Upload clothing flats, select an AI model, and generate realistic, professional styled photos instantly — ready for your product pages.

"Is expensive studio photography delaying your product launches and eating into your margins?"

SupabaseNano Banana AIWeb Interface
The Problem
📸

Expensive photoshoots

Every new garment required a model, photographer, studio, and stylist — high cost and weeks of lead time before anything could be listed online.

⏳

Slow time-to-market

New inventory sat in the warehouse while waiting for shoot slots. By the time photos were ready, the trend window had often passed.

🔄

No flexibility

Once shot, changing a garment on a different model or in a different style meant starting the entire process over — no easy way to iterate.

How It Works
01

Garment Uploaded

Brand uploads clothing items — jeans, tops, jackets, and other wearables — via the web interface.

02

Model Selected

User picks from pre-trained AI model profiles stored in Supabase.

03

AI Renders

The garment-model combination is processed through the Nano Banana AI engine.

04

Photo Delivered

A professional styled model photo is returned in seconds — ready to publish.

Tech Stack
Web Interface

Styling Interface

A clean browser-based tool for uploading garments, selecting models, and managing outfits. No software to install — works entirely in the browser.

  • Garment upload
  • Model selection
  • Real-time swap & preview
Supabase

Data Layer

Stores all uploaded garments and pre-trained AI model profiles. Manages garment state and enables dynamic layering, swapping, and outfit control.

  • Garment storage
  • Model profile library
  • Dynamic outfit state
Nano Banana AI

AI Rendering Engine

Processes each garment-model combination and generates the final styled output photo. Handles realistic draping, lighting, and fit across a wide range of garment types.

  • Realistic garment rendering
  • Instant output generation
  • Multiple garment types
Live Example
📤 summer_denim_jacket.jpg uploaded
👤 Model: Profile #3 — Female, Slim fit
🎨 Style: Casual outdoor
▶ Generate clicked
→
✓ Model photo generated in 4.2s
✓ Realistic drape & fit applied
✓ Ready to download for product listing
No photoshoot needed
Key Features
⚡

Instant Photo Generation

Professional styled model photos generated in seconds — no scheduling, no studio, no wait.

👗

Multiple Garment Types

Works with jeans, tops, jackets, and other wearables — handles realistic draping and fit for each category.

🔄

Real-Time Swapping

Garments can be swapped, layered, or removed in real time — see the result instantly without regenerating from scratch.

🧍

Pre-Trained AI Models

A library of model profiles stored in Supabase — different body types, styles, and aesthetics to match any brand direction.

💰

Dramatically Lower Cost

Eliminates the cost of photographers, models, studios, and stylists — the same output at a fraction of the price.

🚀

Fast E-Commerce Updates

New inventory can be listed with professional photos the same day it arrives — no waiting for shoot slots.

List new inventory
the day it arrives.

Built for fashion brands and e-commerce teams. Custom model library, any garment type, live in days.

  • Custom AI model library
  • Any garment category
  • Live in days, not months