AI Engineer | Building AI Agents, Chatbots & Automations
New to Contra
AI Engineer | Building AI Agents, Chatbots & Automations
๐๐๐ฟ๐ฎ๐น๐ถ๐ ๐๐ โ ๐ข๐ป๐ฒ ๐๐ป๐๐ฒ๐น๐น๐ถ๐ด๐ฒ๐ป๐ ๐๐ด๐ฒ๐ป๐. ๐ ๐๐น๐๐ถ๐ฝ๐น๐ฒ ๐๐๐๐ถ๐ป๐ฒ๐๐ ๐ฆ๐ผ๐น๐๐๐ถ๐ผ๐ป๐.
What if your business had one intelligent assistant helping you manage customer conversations, organize leads, coordinate tasks, and reduce repetitive work?
That's the vision behind ๐๐๐ฟ๐ฎ๐น๐ถ๐ ๐๐, a concept for an AI-powered business agent designed to bring everyday operations into one connected workspace.
๐ง๐ต๐ฒ ๐ฃ๐ฟ๐ผ๐ฏ๐น๐ฒ๐บ
Businesses often manage customer messages, leads, CRM updates, scheduling, and documents across multiple disconnected tools.
This ๐ฐ๐ฟ๐ฒ๐ฎ๐๐ฒ๐ ๐ฟ๐ฒ๐ฝ๐ฒ๐๐ถ๐๐ถ๐๐ฒ ๐๐ผ๐ฟ๐ธ, ๐ถ๐ป๐ฐ๐ผ๐ป๐๐ถ๐๐๐ฒ๐ป๐ ๐ถ๐ป๐ณ๐ผ๐ฟ๐บ๐ฎ๐๐ถ๐ผ๐ป, ๐ฎ๐ป๐ฑ ๐๐ป๐ป๐ฒ๐ฐ๐ฒ๐๐๐ฎ๐ฟ๐ ๐ฎ๐ฑ๐บ๐ถ๐ป๐ถ๐๐๐ฟ๐ฎ๐๐ถ๐๐ฒ ๐ผ๐๐ฒ๐ฟ๐ต๐ฒ๐ฎ๐ฑ.
๐๐๐ฟ๐ฎ๐น๐ถ๐ ๐๐ explores a simpler approach: one coordinated system that helps people and business applications work together.
๐ง๐ต๐ฒ ๐ฃ๐ฟ๐ผ๐ฝ๐ผ๐๐ฒ๐ฑ ๐ฆ๐ผ๐น๐๐๐ถ๐ผ๐ป
๐๐๐ฟ๐ฎ๐น๐ถ๐ ๐๐ is designed around six core capabilities:
1) ๐๐ ๐๐๐๐๐ผ๐บ๐ฒ๐ฟ ๐ฆ๐๐ฝ๐ฝ๐ผ๐ฟ๐ โ Assist with common inquiries and customer conversations.
2) ๐๐ฒ๐ฎ๐ฑ ๐ค๐๐ฎ๐น๐ถ๐ณ๐ถ๐ฐ๐ฎ๐๐ถ๐ผ๐ป โ Capture inquiries, organize prospect information, and support follow-up decisions.
3) ๐๐ฅ๐ ๐๐๐๐ผ๐บ๐ฎ๐๐ถ๐ผ๐ป โ Connect customer interactions with CRM updates and lead-management workflows.
4) ๐ฆ๐ฐ๐ต๐ฒ๐ฑ๐๐น๐ถ๐ป๐ด & ๐ฅ๐ฒ๐บ๐ถ๐ป๐ฑ๐ฒ๐ฟ๐ โ Coordinate appointments, notifications, and recurring tasks.
5) ๐๐ผ๐ฐ๐๐บ๐ฒ๐ป๐ ๐๐ป๐๐ฒ๐น๐น๐ถ๐ด๐ฒ๐ป๐ฐ๐ฒ โ Summarize documents, extract useful information, and assist with information processing.
6) ๐ช๐ผ๐ฟ๐ธ๐ณ๐น๐ผ๐ ๐๐๐๐ผ๐บ๐ฎ๐๐ถ๐ผ๐ป โ Coordinate supported applications and business processes through agent-assisted actions.
Human Control Where It Matters
The proposed system also includes human review and approval for sensitive or consequential actions, helping businesses maintain oversight while benefiting from automation.
๐๐ฒ๐๐ถ๐ด๐ป & ๐ฃ๐ฟ๐ผ๐ฑ๐๐ฐ๐ ๐๐ถ๐ฟ๐ฒ๐ฐ๐๐ถ๐ผ๐ป
I developed the product concept and visual direction around a unified SaaS dashboard, combining customer conversations, agent activity, workflow management, and lead information in one interface.
The design uses an obsidian background, emerald and amethyst accents, layered glass panels, and structured information hierarchy to communicate a premium enterprise AI experience.
๐ง๐ต๐ฒ ๐๐ถ๐ด๐ด๐ฒ๐ฟ ๐ฉ๐ถ๐๐ถ๐ผ๐ป
My goal with ๐๐๐ฟ๐ฎ๐น๐ถ๐ ๐๐ is to explore how businesses could move from disconnected manual processes toward one coordinated, intelligent system.
Less repetitive work. More organized operations. More time to focus on growth.
2
134
Veyra โ AI Lead Qualification & CRM Automation
One Intelligent Platform. Multiple Business Solutions.
Managing leads, qualifying potential customers, updating CRM records, and following up with prospects can consume valuable time and resources. Businesses often rely on multiple tools and employees to handle these repetitive tasks.
Veyra is designed to bring everything together in one AI-powered workspace.
What Veyra Aims to Do
AI Lead Qualification: Analyze incoming leads, assess their potential, and recommend which opportunities deserve attention.
Smart CRM Management: Organize customer information, track sales pipelines, and manage opportunities in one place.
Automated Follow-Ups: Generate personalized messages, schedule reminders, and help prevent missed opportunities.
AI-Powered Workflows: Coordinate lead capture, qualification, CRM updates, and follow-up tasks with human approval.
Business Insights: Provide dashboards that help businesses understand their sales activities and prioritize next steps.
The Goal
To help businesses reduce repetitive administrative work, simplify sales operations, and manage more opportunities without constantly switching between different tools.
Veyra is more than a CRM concept โ it's a vision for an intelligent sales operations assistant that supports teams from the first lead to the next business opportunity.
1
129
Relay โ AI Customer Support & Lead Capture
One intelligent platform. Multiple business solutions.
Running a business means handling customer inquiries, answering repetitive questions, managing leads, and following up with potential clients. These tasks often require multiple employees, tools, and valuable time.
Relay aims to bring everything together in one intelligent platform.
Designed to act as a 24/7 AI-powered business assistant, Relay helps businesses:
Automate Customer Support โ Answer common questions without constant manual assistance.
Capture Potential Clients โ Turn customer conversations into organized business leads.
Manage Conversations โ Keep customer interactions and inquiries in one workspace.
Enable Human Handoff โ Let staff take over conversations whenever needed.
Streamline Operations โ Reduce repetitive administrative work and the need for separate tools.
The vision: One unified AI solution that helps businesses manage customer interactions, reduce operational workload, and grow without unnecessarily expanding their support team.
1
148
๐ฌ๐ผ๐๐ฟ ๐๐๐๐ถ๐ป๐ฒ๐๐ ๐๐ผ๐ฒ๐๐ป'๐ ๐ก๐ฒ๐ฒ๐ฑ ๐ ๐ผ๐ฟ๐ฒ ๐ ๐ฎ๐ป๐๐ฎ๐น ๐ช๐ผ๐ฟ๐ธ. ๐๐ ๐ก๐ฒ๐ฒ๐ฑ๐ ๐ฆ๐บ๐ฎ๐ฟ๐๐ฒ๐ฟ ๐ฆ๐๐๐๐ฒ๐บ๐.
Think about how much time you spend every week doing the same things.
Replying to customer emails.
Following up with potential clients.
Updating spreadsheets.
Managing leads.
Moving information between applications.
Sending reminders and notifications.
Now ask yourself:
๐๐ผ๐ ๐บ๐๐ฐ๐ต ๐ผ๐ณ ๐๐ต๐ถ๐ ๐ฎ๐ฐ๐๐๐ฎ๐น๐น๐ ๐ป๐ฒ๐ฒ๐ฑ๐ ๐๐ผ ๐ฏ๐ฒ ๐ฑ๐ผ๐ป๐ฒ ๐บ๐ฎ๐ป๐๐ฎ๐น๐น๐?
Many repetitive business tasks can be automated, yet businesses continue spending valuable hours doing them manually.
๐ง๐ต๐ฒ ๐๐ผ๐ฝ 1% ๐บ๐ถ๐ป๐ฑ๐๐ฒ๐ ๐ถ๐๐ป'๐ ๐ฎ๐ฏ๐ผ๐๐ ๐๐ผ๐ฟ๐ธ๐ถ๐ป๐ด 24 ๐ต๐ผ๐๐ฟ๐ ๐ฎ ๐ฑ๐ฎ๐. ๐๐'๐ ๐ฎ๐ฏ๐ผ๐๐ ๐ฏ๐๐ถ๐น๐ฑ๐ถ๐ป๐ด ๐๐๐๐๐ฒ๐บ๐ ๐๐ต๐ฎ๐ ๐บ๐ฎ๐ธ๐ฒ ๐๐ผ๐๐ฟ ๐ฏ๐๐๐ถ๐ป๐ฒ๐๐ ๐บ๐ผ๐ฟ๐ฒ ๐ฒ๐ณ๐ณ๐ถ๐ฐ๐ถ๐ฒ๐ป๐.
Imagine a workflow where:
โข A customer submits an inquiry, and their details are automatically recorded.
โข Incoming emails are categorized, and relevant replies are drafted for review.
โข New leads are added to your CRM without manual data entry.
โข Important tasks trigger notifications and reminders automatically.
โข Your business applications exchange information without constant copying and pasting.
โข AI helps process documents, organize information, and handle repetitive administrative work.
๐๐ถ๐ณ๐ณ๐ฒ๐ฟ๐ฒ๐ป๐ ๐๐ฎ๐๐ธ๐. ๐๐ถ๐ณ๐ณ๐ฒ๐ฟ๐ฒ๐ป๐ ๐ฎ๐ฝ๐ฝ๐น๐ถ๐ฐ๐ฎ๐๐ถ๐ผ๐ป๐. ๐ข๐ป๐ฒ ๐ฐ๐ผ๐ป๐ป๐ฒ๐ฐ๐๐ฒ๐ฑ ๐ฎ๐๐๐ผ๐บ๐ฎ๐๐ถ๐ผ๐ป ๐๐๐๐๐ฒ๐บ.
That's what I help businesses build.
I've launched my new service on Contra:
๐๐๐๐๐ผ๐บ ๐๐ ๐๐๐๐ผ๐บ๐ฎ๐๐ถ๐ผ๐ป ๐ช๐ผ๐ฟ๐ธ๐ณ๐น๐ผ๐๐ ๐ณ๐ผ๐ฟ ๐ฌ๐ผ๐๐ฟ ๐๐๐๐ถ๐ป๐ฒ๐๐
I build workflows using n8n, Python, and APIs to connect business applications, reduce repetitive work, and simplify daily operations.
My approach is simple:
โข Understand your existing business process.
โข Identify the repetitive tasks taking up your time.
โข Build an automation workflow around your actual needs.
โข Test the workflow and make sure it performs as intended.
๐๐๐๐ผ๐บ๐ฎ๐๐ถ๐ผ๐ป ๐๐ต๐ผ๐๐น๐ฑ๐ป'๐ ๐บ๐ฎ๐ธ๐ฒ ๐๐ผ๐๐ฟ ๐ฏ๐๐๐ถ๐ป๐ฒ๐๐ ๐บ๐ผ๐ฟ๐ฒ ๐ฐ๐ผ๐บ๐ฝ๐น๐ถ๐ฐ๐ฎ๐๐ฒ๐ฑ. ๐๐ ๐๐ต๐ผ๐๐น๐ฑ ๐บ๐ฎ๐ธ๐ฒ ๐ฟ๐๐ป๐ป๐ถ๐ป๐ด ๐ถ๐ ๐ฒ๐ฎ๐๐ถ๐ฒ๐ฟ.
You don't have to automate everything overnight. Sometimes, solving just one repetitive problem is the best place to start.
Your time should go toward growing your business, not repeating tasks that a system could handle.
Explore my service:
https://contra.com/s/2PUqmBeN-i-will-build-an-ai-automation-workflow-for-your-business
Business owners and founders: What's one repetitive task you wish you could automate today?
#AIAutomation #BusinessAutomation #WorkflowAutomation
FinValora โ Real-Time FinTech & Paper-Trading Engine
Project Overview
FinValora is a modular, high-throughput cryptocurrency paper-trading and portfolio management backend. Built to simulate high-frequency financial platforms, it pairs a transactional Django / DRF foundation for state management with FastAPI for low-latency market streaming, offloading computational workloads to Celery & Redis.
Roles & Tools
Roles: Backend Architect, API Engineer, System Designer
Tools: Python, FastAPI, Django, Django REST Framework, WebSockets, Redis, Celery, PostgreSQL, Docker, Pytest, OpenAPI/Swagger
Core Highlights & Architecture
Hybrid Async Architecture
Transactional API (Django/DRF): Handles secure JWT auth (SimpleJWT), user accounts, wallets, order books, and multi-asset portfolio logic.
Real-Time Streaming (FastAPI + WebSockets): Dedicated async microservice for live price tickers, market data streaming, and WebSocket broadcasts over Redis Pub/Sub.
High-Performance Background Processing
Integrated Celery + Redis task queues to handle heavy financial math, automated portfolio snapshot calculations, equity curves, and drawdowns.
Reduced core HTTP response times by ~25% by executing asynchronous alert triggers and notification pipelines in background workers.
Dedicated Analytics Engine
Isolated data system for trade statistics, historical indicator calculations, strategy performance evaluation, and daily snapshot tracking.
Developer Experience & Quality Assurance
Interactive API documentation generated dynamically via drf-spectacular (OpenAPI / Swagger UI).
Unit & integration test suites validating transactional integrity, risk limits, and WebSocket handshakes.
Links & Resources
GitHub Repository: https://github.com/devHashim111/finvalora
Here is a high-converting, concise, and interactive Contra project post structured to showcase your architectural skills without wall-of-text fluff.
FinValora โ Real-Time Crypto Trading & Portfolio Engine
FinValora is a modular, backend-heavy financial platform built to demonstrate high-concurrency microservices, real-time data streaming, and robust financial domain modeling.
Core Architecture & Stack
Transactional Core: Django & Django REST Framework (DRF) + SimpleJWT
Real-Time Engine: FastAPI + Async WebSockets (isolated for high throughput)
Background Processing: Celery + Redis (notifications, scheduled tasks, market sync)
API Specs & Docs: OpenAPI 3.0 / Swagger UI via drf-spectacular
Data Layer: Relational schema optimized for PostgreSQL (SQLite for local dev)
Key Technical Highlights
Financial Domain Modeling: Strict relations handling assets, exchanges, order books, trades, paper-trading wallets, and positions.
Separation of Concerns: Sync REST APIs handle transactional metadata, while async FastAPI pipelines stream live market feeds via WebSockets.
Performance & Analytics: Dedicated analytics layer computing daily equity curves, trade stats, drawdown history, and strategy metrics.
Notification System: Multi-channel queue pipeline (Email, In-App, Push) executed out-of-band via Redis & Celery workers.
Which part of the architecture interests you most?
Asynchronous WebSocket streaming via FastAPI & Redis Pub/Sub
Database design for order books & portfolio analytics
Decoupling long-running tasks with Celery background workers
Drop a comment below with your pick, or check out the code repositories!
Links
GitHub Repository: https://github.com/devHashim111/finvalora
Developer Portfolio: https://devhashim.vercel.app/
0
68
Project Title
Async E-Learning Platform Backend
Overview
An asynchronous REST API and real-time WebSocket backend built with FastAPI, Fapix, and Tortoise ORM. Features strict Role-Based Access Control (RBAC), live multi-room chat, assessment workflows, and an automated async test suite.
Key Deliverables
Role-Based Security: Built fine-grained RBAC permission guards across 4 user roles (Admin, Supervisor, Teacher, Student).
Real-Time Communication: Designed multi-room WebSocket chat channels using structured JSON payload envelopes.
Database & Testing: Implemented async data persistence with Tortoise ORM and built an automated test suite using pytest-asyncio and httpx.
Live Demo & Source Code
Live App: https://lms-4475.de (https://lms-4475.deCould)
Source Code:https://github.com/devHashim111/E-Learning-System
2
235
Fapix โ Open Source Asynchronous FastAPI Toolkit
Role: Creator & Lead Maintainer | Active Development
CLI & Scaffolding: Built CLI commands using Typer to instantly bootstrap modular FastAPI architectures with pre-configured directory structures, settings, and database integrations.
Schema & Lifecycle Hooks: Designed action-specific request/response schema helpers and automated startup/shutdown ORM hooks to streamline data validation and connection management.
Real-time & Async Support: Included utility abstractions for managing WebSocket connections, asynchronous background tasks, and lightweight API routing defaults.
Packaging & Distribution: Maintained package builds, versioning, and distribution on PyPI for seamless pip install integration.
Mail Hauler Pro Email Marketing & Automation SaaS
Developed Mail Hauler Pro, an enterprise-grade email marketing and automation platform built to support 12,400+ marketing teams.
We created the complete SaaS product from campaign management and multi-inbox operations to email deliverability monitoring and real-time performance analytics. The platform helps teams manage high-volume outreach, automate campaigns, protect sender reputation, and handle replies from one unified workspace.
Key features:
โข Automated email campaigns and sequences
โข Multiple sending-account management
โข Automated email warm-up
โข Unified inbox for managing replies
โข Lead import and subscriber management
โข Lead generation workflows
โข Bounce and deliverability monitoring
โข Sender health tracking
โข Real-time campaign analytics
โข Infrastructure for high-volume email operations
Technology:
React, Next.js, Node.js, MongoDB
Status:
Live and scaling Private Beta v1.0
2
102
Most 404 pages say "oops." Ours runs a terminal simulation.
The Goal: Turn a bounce into a "wow" moment.
The Strategy: Treat every error as an Easter egg.
The Result: A dead end becomes the most shared page on the site.
Every pixel is an opportunity to prove you're better than the "standard" version.
Link in bio if you want this for your project.
Exploring structured prompting with Kittl.
For this entry, I created a cinematic luxury footwear concept that captures the moment just before completion, where each component comes together to form the final design.
The project includes a womenโs high heel video and a three-image sneaker composition, focusing on detail, craftsmanship, and premium aesthetics through realistic textures and refined branding.
This highlights how structured prompting can transform an idea into a visually polished result.
1
211
๐ฅ Full-Stack Development - RankBee AI
I worked as a Full-Stack Developer on RankBee an AI visibility platform that helps brands track and improve how they appear across AI search engines.
๐ก Key contributions: โก Interactive dashboards for AI visibility โก Tools to track competitors and benchmark presence โก AI-driven content optimization workflows โก Prompt-based testing systems โก Streamlined data processing into actionable insights
The platform shows how AI interprets a brand which competitors are recommended and how to optimize content for better AI visibility.
I focus on solving complex technical challenges and delivering products that give businesses a real advantage. If youโre developing AI-driven tools or ambitious platforms, letโs collaborate.
2
2
244
I also designed and developed https://infinite-compute.com/, a cloud-powered platform that enables users to access high-performance GPU machines and run demanding 3D tools directly from the web without expensive local hardware. I built the complete system architecture myself including the frontend backend services and cloud infrastructure focusing on performance reliability and secure access to remote computing resources. The platform provides a responsive interface to launch and manage GPU sessions while the backend handles session orchestration user management and compute allocation. The system was designed to maintain stable remote sessions efficiently manage resources and provide secure authentication so developers designers and 3D artists can run heavy software on demand from anywhere.
Multi-Agent AI System for Intelligent Workflow Automation
Designed and developed a multi-agent AI system featuring specialized agents for sales, technical delivery, marketing, and operations. The project focuses on coordinating intelligent agents to automate workflows, handle specialized tasks, and improve collaboration between different functional areas.
Built using Python, Google ADK, Model Context Protocol (MCP), and FastAPI, with a React.js interface for interacting with the system. Docker and GitHub supported containerization, development workflows, and project management.
The project combines agentic AI, backend engineering, and frontend development to create a modular system that can be extended with new agents, tools, and capabilities.
0
73
Auto Electronics โ SMD Digital Advertising Display Solution
The Auto Electronics Project focuses on digital advertising and display solutions powered by SMD (Surface-Mounted Device) screens. The project aims to help businesses present high-quality visual advertisements and promotional content in public and commercial spaces.
My work involved contributing to the projectโs technical development, focusing on web application functionality, interface design, debugging, and problem-solving to support the overall digital display solution.
The project combines digital display technology with software development to create practical solutions for presenting engaging visual content. The goal is to improve how businesses communicate promotions, showcase products, and deliver advertising messages through modern digital screens.
0
69
RedVapt โ Agentic AI Penetration Testing Framework
RedVapt is an agentic AI-powered penetration testing framework designed to streamline security assessments through intelligent automation and collaborative AI agents. The platform focuses on automating reconnaissance, security scanning, and vulnerability analysis while providing an interface for managing security assessments and reviewing findings.
The project combines AI-driven workflows with web application development to help organize security testing tasks, analyze potential vulnerabilities, and make assessment results easier to understand.
Built with a focus on intelligent automation, problem-solving, and user experience, RedVapt explores how agentic AI can support modern cybersecurity workflows and reduce repetitive manual effort.
The framework is intended for authorized security testing and controlled environments, with an emphasis on responsible vulnerability assessment.
Developed a full-stack language learning application tailored for Luxembourgish, combining speech recognition, natural language understanding, and generative AI. Fine-tuned OpenAIโs Whisper model for accurate Luxembourgish transcription and built a custom text-to-speech (TTS) engine for realistic audio feedback. A RAG-based architecture enables the app to answer user queries contextually, making learning highly interactive. The frontend is built with React, while Flask powers the backend. Designed to deliver an immersive, conversation-driven auditory learning experience.
1
183
Hey Contra!
Excited to finally be here and connect with builders, founders, and innovators from around the world.
I'm Osman, an AI Engineer specializing in AI Agents, LLM Applications, Automation Systems, and Full-Stack Development. Over the past 3+ years, I've helped businesses transform ideas into intelligent products using OpenAI, Gemini, LangChain, LangGraph, FastAPI, Django, and modern cloud technologies.
Some of the solutions I've built include:
โข AI Agents and Multi-Agent Systems
โข RAG Applications with Private Knowledge Bases
โข AI-Powered SaaS Products
โข Automation Workflows with n8n, Zapier, and Make
โข Voice AI Agents with Twilio and ElevenLabs
โข Custom Chatbots and Internal AI Tools
โข Data Automation and Web Scraping Systems
I enjoy solving complex problems and building products that create real business valueโnot just demos.
Looking forward to connecting with founders, startups, agencies, and teams working on exciting AI projects.
If you're building something ambitious with AI, let's talk.
1
144
Retail Knowledge Graph
In this project, we built a semantic knowledge graph tailored to the retail industry. The pipeline involved developing AI agents to transform heterogeneous data into standardized formats. Ontologies were created to represent domain knowledge accurately. Using Gemini models and LangChain, user queries were converted into Cypher queries to retrieve insights from a Neo4j database. We utilized an MCP server for orchestration and LangSmith for secure login and audit trails. This system enhances complex data exploration for non-technical users.
2
2
208
Student Medical Chatbot
Built a chatbot to assist MBBS students in navigating medical literature. Leveraged Llama Index and fine-tuned language models to ensure accuracy. Embeddings were stored in OpenSearch, hosted on AWS. The Django backend included secure authentication and session management for a robust user experience.