Partner
Projects
PaidT - Price Action Intelligent Detection Trading
Introduction
PaidT is a software system for automated trading based on recognizing patterns in price changes. The system also supports simulation and emulation of trading outcomes in realistic conditions.
https://paid-t.comClient requirement
The client needed support in developing and setting up the complete software system for scalable data analysis and high-performance decision-making.
How did INVT address the challenge
The team developed the backend system for data analysis, trading simulation, and execution. Scalable cloud hosting and failover mechanisms were put in place. Complete technical project management is provided. Supporting the client in shaping the product roadmap and exploring new technologies.
Used technologies
Python, Pytrader, Pandas, EightCap, PostgreSQL, Google Cloud VM
Bundlr - Digital Content Aggregation Platform
Introduction
Bundlr is a software platform for aggregation of digital content from different sources, including subscription-based services. The solution applies AI for content annotation and advanced search based on user profile setup and custom prompts.
https://www.bundlrco.com/Client requirement
The client needed support for developing and integrating the whole solution, including the backend services, AI components, scalable hosting, and mobile app implementations.
How did INVT address the challenge
The team performed enhancement of the existing and implementation of new backend services. Scalable and high-performance cloud setups are realized. The team implemented Bundlr mobile application using cross-platform frameworks. Full technical management of the project is provided.
Used technologies
Python, FastAPI, PostgreSQL, Flutter Dart, ElasticSearch, Docker, Google Cloud, Bitrise, Whisper, Terraform, Github Actions
Business Intelligence SaaS
Introduction
SaaS for advanced business process intelligence and management.
Client requirement
Support the existing engineering team with frontend development services.
How did INVT address the challenge
INVT’s frontend developers joined the client's team and performed a full set of frontend development services. CI/CD for the frontend component of the solution is set up and maintained.
Used technologies
React, Azure AD B2C, Azure Storage, Github Actions, BPMN, React Query, Github Packages
Website for TV Station
Introduction
Client is a local TV station and news outlet.
Client requirement
The client approaches INVT with requirements for developing and hosting website for the TV channel and news outlet.
How did INVT address the challenge
In line with the project requirements and limitations, the team opted for WordPress template and website hosting. The team performs monthly maintenance and monitoring of the deployed website.
Used technologies
WordPress
Know Your Customer SaaS
Introduction
The project is a SaaS for financial institutions for streamlining the client onboarding, collection of mandatory KYC information and documents and organizing due diligence and regulatory compliance tasks.
Client requirement
The client needed a proof of concept implementation of the service with integration with a preselected document management database.
How did INVT address the challenge
The team researched the challenges in KYC and how digital systems can support the process. The enterprise document management software was investigated. The proof of concept is implemented for demonstration purposes including the backend services, AI-based document analysis and annotation services, integration with the document management system and frontend application based on custom design.
Used technologies
Python, ReactJS, Mobius Database, FastAPI, PostgreSQL
AI Chatbot for Energy Exchange
Introduction
The client provides a web platform which brings together energy suppliers, retailers and brokers.
Client requirement
The client approached the INVT team with requirements to build a proof of concept for an AI powered chatbot which will be able to map typical user prompts onto proper system operations and API calls.
How did INVT address the challenge
The team performed detailed inspection of the platform’s APIs and capabilities and was briefed into the basics of energy retailing in the US market. The team integrated OpenAI APIs with custom Python scripts for text analysis. Everything was presented in a chatbot popup and integration was tested with a selected platform’s APIs.
Used technologies
Vue.js, Python, Open AI
Agile Management for eSports Tournament Platform
Introduction
Client is an esports tournament platform, building efficient payment infrastructure to enable gamers to earn and achieve. It is based on web3 and blockchain technologies.
Client requirement
The client required an experienced scrum master with a good technological background which could improve efficiency of agile software development processes across multiple remote teams.
How did INVT address the challenge
A INVT team member joined the client’s teams and streamlined agile software processes and established an effective set of scrum practices.
Used technologies
Agile, Scrum, Jira, JRC
Digital System For Alternative Investment Fund
Introduction
Client provides alternative investment funds. They provide web tools to the investors for portfolio management.
Client requirement
The client requested a custom website implementation with a specific set of services for managing the register of investment units for alternative investment funds. The local regulation had to be fulfilled.
How did INVT address the challenge
The team researched the domain of alternative investment funds and governing regulations and set of requirements defined for the funds and how they must keep record of the investment units. An extended design and development team was established. The complete website and supporting backend services were designed, implemented, integrated and tested.
Used technologies
PHP, HTML, CSS, Apache Server, MySQL
Embedded System Test Tool
Introduction
The client was responsible for building embedded software and a support application for a system for rapid fire suppression in gas lines.
Client requirement
The client required updates for the system’s support app which would enable verification of firmware upload process for the field devices, monitoring firmware upload process with a failover mechanism.
How did INVT address the challenge
The team got familiar with the existing software stack and services and specified necessary additions. New software logic was implemented for integration with the existing application. Tests are performed on the actual field device.
Used technologies
C++
CyberGRID - Virtual Power Plant Solution
Introduction
CyberGRID is a Virtual Power Plant solution for seamless flexibility management and monetization.
https://www.cyber-grid.com/Client requirement
Clients require a software solution for managed distributed power flexibility that aggregates their diverse energy assets - such as batteries, renewables, and industrial loads - into an optimized Virtual Power Plant (VPP). This enables them to monetize their flexibility by actively participating in various electricity markets. The primary goals are to unlock new revenue streams and enhance the power grid's overall stability.
How did INVT address the challenge
The team implemented new services and advanced algorithms, executed a complete redesign of the user interface (UI) to enhance user experience, and successfully integrated the platform with third-party services. Furthermore, the team dedicated its efforts to optimizing and significantly improving existing functionalities.
Used technologies
Java, PostgreSQL, Angular
Forkast - Web3 Prediction Platform
Introduction
Forkast.gg is a Web3 prediction market platform for gaming, esports, and internet culture.
https://forkast.gg/Client requirement
INVT designed a deep research, multi-hop chatbot with strict behaviour guardrails that fetches and summarises recent stories and ongoing developments from multiple, opposing views to present a user with the most relevant, unbiased information.
How did INVT address the challenge
Teaching a cheap model to use proper information-collection tactics through multi-shot in-context learning.
Used technologies
FastAPI, Google AI Studio, Langchain
Upstak - AI-Powered CRM & Workflow Automation
Introduction
Upstak is a Value-Added Reseller specializing in Data Center Infrastructure Products. The project's goal is to create a sophisticated AI assistant to support Upstak's internal employees. Hal automates complex CRM operations, answers data-driven questions, and retrieves information from a knowledge base, streamlining workflows and improving efficiency.
https://www.upstak.com/Client requirement
The primary challenge is automating complex, multi-step, and error-prone business processes within Upstak's Method CRM. Manually creating records like Estimates or Opportunities involves multiple dependencies (customers, items, pricing, margins) and requires significant time and attention to detail. All this needs to be available in a web application and integrated with Slack and Mail.
How did INVT address the challenge
INVT designed a sophisticated, multi-agent architecture using LangGraph to address the complexity of the requirements. Instead of a single monolithic chain, the system delegates tasks to specialized agents, each modeled as a state machine.
Used technologies
Python, PostgreSQL, LangChain, LangGraph, FastAPI
Coinsub - Crypto Subscription Management
Introduction
Coinsub is a software platform designed to help merchants manage subscription-based payments and complex submerchant structures. Coinsub enables one-time and subscription-based businesses to accept crypto payments, specifically stablecoins, in a seamless, non-custodial, and trustless manner.
https://coinsub.io/Client requirement
The Coinsub team requested help with the day-to-day full-stack. The client required a secure and scalable full-stack platform to handle recurring subscription payments for a multi-merchant environment. Key needs included a robust API for merchants, detailed financial reporting, a secure system, and blockchain software development.
How did INVT address the challenge
The team developed the full-stack application, including a high-performance backend API using Go and server-side rendering, a responsive merchant dashboard, a buy app, and a mobile app.
Used technologies
Go (Golang), Gin, PostgreSQL, Goose, Tailwind CSS, JavaScript, Pusher