WhatsApp lead follow-up
A new enquiry comes in, AI reads it, tags the lead and sends the right reply or a payment link. Nobody has to watch the inbox.
AI Automation ExpertDelhi, India
AI Automation ExpertFull Stack DeveloperFlutter DeveloperGraphic Designer & Video EditorFounder & CTO, Slackroid
I automate the boring parts of a business with AI, and build the apps and websites that run the rest.
Teams I've worked with
01 / About
I'm an AI automation engineer and full stack developer from Delhi. In 2019 I started Slackroid, and since then I've built most of its products myself: WhatsApp automation, payment flows, Flutter apps, e-commerce stores and the Linux servers they run on.
My favourite work is automation. I wire WhatsApp, payment gateways, CRMs and LLMs together so that replies, follow-ups, invoices and reports happen without anyone touching them. At Radhey Radhey Communication I did this for daily operations, and I've integrated 20+ APIs across client projects.
Since 2024 I've also worked with AI companies like Aligned Labs, Alignerr, Outlier and Soul AI, reviewing more than 1,000 model responses for safety, accuracy and reasoning. That work shapes how I build with AI: test it properly, and don't trust output you haven't checked.
I also design. At Newtech Video Systems I made marketing creatives, exhibition graphics and product videos, and audited the website for SEO. So I can take a project from the database to the launch video.


Founder & CTO · Slackroid · since 2019
"We partnered with Slackroid for custom software development, and the results exceeded expectations. Their team combined technical expertise with clear communication."
"We improved our retention rates after adopting Slackroid's feedback tool. It helped us act quickly on customer needs."
"The automation features alone have saved us countless hours, and the analytics provide insights we never had access to before."
02 / Core expertise
If someone on your team copies data between apps, sends the same reply ten times a day, or chases payments by hand, that's a job for automation.
Replies get drafted by AI, leads get routed, invoices and reminders go out, and reports show up every morning. It's the same stack that runs Slackway.
A new enquiry comes in, AI reads it, tags the lead and sends the right reply or a payment link. Nobody has to watch the inbox.
When Razorpay or Cashfree confirms a payment, the order is updated, the invoice goes out and the customer gets a WhatsApp receipt.
Recharges, bills and renewals are tracked in the database and reminders go out automatically, like in NumStock.
Sales, leads and pending tasks are collected overnight and land as a short summary on WhatsApp or email every morning.
Common questions get a drafted answer from an LLM, checked against your own data, so the team only edits and sends.
Backups, uptime checks, DNS and deploy steps scripted on Linux so they run on schedule instead of from memory.
Also · Full stack & Flutter
Flutter apps, PHP/MySQL APIs and dashboards, with payments, push notifications and WhatsApp alerts built in. Android & iOS from one codebase, Razorpay / Cashfree checkout, deployed on hardened Linux servers.
03 / Services
Workflows across WhatsApp, CRMs, payments and reports, so routine replies, follow-ups and paperwork happen without anyone touching them.
↗Mobile apps in Flutter with PHP/MySQL backends, push notifications and in-app payments, designed from the database up.
↗Storefronts, seller panels and checkout with Razorpay, Cashfree and UPI, from a single shop to a multi-vendor marketplace.
↗Prompt design, rubric-based output review and evaluation workflows for international AI labs since 2024.
↗Marketing creatives, exhibition graphics, product videos and short reels, plus website and SEO audits.
↗04 / How I work
Sit with the people who'll use it and find the steps they repeat every day.
Design the database and APIs first, so the app, payments and reports agree.
Flutter, PHP, Python and MySQL, wired to WhatsApp, Razorpay, Cashfree or an LLM as needed.
Deploy on hardened Linux servers and keep it running after launch.
05 / Licenses & certifications
Each card says what the course covered and where I use it. Open any card for the original certificate.
A hands-on workshop on using AI tools for presentations, data analysis and coding and debugging. Practical tricks I now use day to day.
View certificate ↗
The longest course here at 13+ hours: service boundaries, API gateways, sagas, CQRS and resilience patterns. It changed how I split Slackroid's payment, messaging and store code into services.
View certificate ↗
How to design dense networks, pick loss functions and optimisers, and spot overfitting with validation data and callbacks. It helps me judge model behaviour in evaluation work, not only use it.
View certificate ↗
Convolutional networks for images, data augmentation, and transfer learning from pre-trained models. Useful when a client asks for things like product-photo sorting and I need to know what's realistic.
View certificate ↗
Tokenising text, word embeddings and sequence models (RNNs and LSTMs) for classification. This connects straight to my audio NLP and LLM evaluation work, and to classifying WhatsApp messages by intent.
View certificate ↗
Windowing time-series data and forecasting with neural networks. I've found it handy for sales and stock trend ideas in the dashboards I build.
View certificate ↗
Using AI on logs and metrics to catch anomalies and automate incident handling. It lines up with how I run servers: monitor, alert, and script the fix.
View certificate ↗
A clear map of the field: machine learning, deep learning, generative AI and where each fits. I lean on it when explaining to clients what AI can and can't do for them.
View certificate ↗
How voice and chat assistants handle intent, context and replies. Directly useful for the WhatsApp bots I build.
View certificate ↗
Picking AI use cases that pay off, getting data ready, and rolling out in stages. This is how I scope automation projects: start with the task that wastes the most hours.
View certificate ↗
Writing infrastructure as code: providers, state, variables and modules. Servers become something you can rebuild from a file instead of from memory.
View certificate ↗
The starting point of the TensorFlow track: setting up the environment, working with tensors, and the basic train, evaluate and save loop in Keras. It's the base I use whenever I prototype a small model or read someone else's TensorFlow code.
View certificate ↗
How the three web languages fit together, and where frameworks come in. A good refresher before rebuilding this portfolio.
View certificate ↗
GitHub Actions, templates and shortcuts that remove manual steps from shipping code. Automation again, just for developers.
View certificate ↗
Repositories, branches, pull requests, issues and reviews. It's how I keep client code versioned and recoverable.
View certificate ↗
Modern JavaScript properly: scope, closures, promises, modules and the DOM. Every dashboard and storefront I've built runs on it.
View certificate ↗
Core Python: data types, functions, modules, files and errors. Python is what I reach for in automation scripts and AI tooling.
View certificate ↗The basics of the iOS toolchain and app lifecycle. It helps when I ship Flutter apps to iPhone and need to sort out the Xcode side.
Guru Gobind Singh Indraprastha University (GGSIPU), Delhi
CGPA 8.6 · 2023 – 2026
Udyam (MSME), GST and DGFT Import Export Code registrations. Founded 2019.
View ↗Public repos and what I'm building now. Most client work is private; happy to walk you through it on a call.
Open GitHub ↗Toolkit
06 / Career
17 roles. Tap a role for details.
07 / FAQ
Mostly AI automation and app development: WhatsApp bots, payment and order flows, reporting, Flutter apps and websites. I also take LLM evaluation projects and design or video work.
Yes. Projects run through Slackroid, my registered company (Udyam, GST), so contracts and invoices are straightforward.
We list the tasks your team repeats every day, pick the one that wastes the most time, and I automate that first. You see it working within the first week or two, then we add the next one.
Reviewing model responses for safety, accuracy, reasoning and alignment, writing and grading coding prompts, testing chat agents like an AI therapy app, and annotating audio for speech models.
Flutter and Dart for mobile, PHP and JavaScript on the backend, MySQL for data, Python for AI and scripts, and Ubuntu/Apache servers. For automation: WhatsApp Cloud API, webhooks, Razorpay, Cashfree and LLM APIs.
Message me on WhatsApp or use the form below. Tell me what you need and by when, and I'll come back with questions or a plan.
08 / Contact
Usually replies within a day. Clear scope, progress you can see every week, and software that keeps running after launch.