Abhijith V K

A web chat interface powered by the latest open source language and image models. It can create custom bots and deploy them across web channels with a JavaScript web component and to other channels such as WhatsApp and WordPress.
The back end is powered by Django, PostgreSQL, Redis, Celery, and Celery Beat, while the front end is written in Vue 3.

This is a knowledge base chat interface for employees and a ticket resolution chat interface with automatic knowledge base creation from past incidents. The team's chat interface was deployed across the web.

A federated machine learning platform with an easy-to-use notebook facilitating custom job execution. The job pipeline is abstracted away from the user, facilitating a platform-agnostic job execution.

It's a platform where you can use interactive UI drag-and-drop items to create a bot to operate as a virtual assistant on platforms such as WhatsApp, Salesforce, Microsoft Outlook, and more. It also supports web SDK to be deployed on any website.

• Restructured the database and application architecture of the bot-builder platform to scale efficiently.
• Developed Python modules to be used by the team for building APIs.
• Optimized the Redis cache layer for memory-efficient storage.
• Reduced the response time with write-through cache with Redis, Celery, and MongoDB.

Created a data extraction pipeline featuring an API-as-a-service architecture, allowing users to build Extract Transform Load (ETL) pipelines from PDFs and images. This facilitates automated data entry across various industries, including accounting software and HTML forms.

SMFactory is a Go package that lets job orchestration. It provides various easy-to-use interfaces for defining jobs and their dependencies, allowing for efficient and reliable execution of complex workflows. SMFactory also supports job parallelism and serializability, making it more versatile and powerful. Additionally, there are plans to develop SMFactory into a CLI tool capable of reading a declarative DSL and running binaries on nodes.

A curated marketplace built from scratch with Instagram-inspired post features for user engagement.

• Connected various third-party APIs for payment, delivery, accounting, mail service, and more.
• Have integrated with Instagram API connection, helping brands integrate their Instagram promotions and posts into the app.
• Developed full-fledged vendor and customer APIs all from scratch.
• Integrated Elasticsearch for main window search on products, brands, tags, and more.

The IoT greenhouse farm project utilizes a Raspberry Pi running a Python client and MQTT as the communication protocol. The system is controlled through a web app, which sends instructions to the Raspberry Pi. The Raspberry Pi's output is then connected to microcontrollers that control the various components of the greenhouse. This allows for remote monitoring and control of the greenhouse, optimizing crop growth and reducing resource waste.

Built a ticketing and trading platform that consists of a Web 3.0 agnostic Web 2.0 app built on the Web 3.0 EVM infrastructure. It has social post features for user traction on Instagram, and the linked posts are imported with one click away for easy ticket creation. Organizations can have new monetization schemes never experienced before.

A real-time weather application with a detailed weather forecast, real-time data, news, alerts, satellite images, and real-time weather maps, all based on geolocation, with various subscription plans. The project name was Metbeat weather, and it is available on both the web and Android.

I was the lead back-end developer for the project, and I developed a database with PostgreSQL in the Django framework. I also developed modules for user subscription plans, news, and alerts, and now cast. I integrated a payment gateway and OpenWeatherMap APIs, which improved the user experience and quality of data displayed. I developed a meta-application and integrated Facebook instant articles to increase the revenue and users from Facebook for the platform.

The application has more than 10 thousand downloads with a 4.3 rating on the Google Play Store.

A cryptocurrency scalp trading bot in Python working on the Binance API with trade decisions based on general trading strategies. I also tried to implement a profit probability distribution approach without much success in the trading outcome. The program was a script running on the terminal for his personal use to trade on his behalf.

Maracc, short for Market Accelerator, is a SaaS platform that enables businesses to automate their social media marketing pipelines from one centralized dashboard supporting all major social media, including Facebook, Instagram, LinkedIn, Google Ads, TikTok, and more. In addition, the platform has an SEO enhancement tool to create AI-powered automated blogs, a hashtag extractor to suggest relevant trending hashtags, an AI-powered automatic title creator, all-in-one place analytics, and more.