I’m a Co-Founder & Chief Product Officer (CPO) at Agus Systems, an AI/ML Engineer, and an active Open Source Contributor.

My expertise spans across Artificial Intelligence, Machine Learning, Large Language Models, SaaS engineering, and automation systems. I specialize in building intelligent systems from the ground up — from training core AI models to deploying production-ready scalable products.

I work at the intersection of:

• AI Research
• Machine Learning Engineering
• Product Architecture
• Automation Systems
• Scalable SaaS Development

My core expertise includes:

• Training custom machine learning models
• Neural network architecture design
• Transformer model fine-tuning
• Data preprocessing and feature engineering
• Model optimization and evaluation
• Hyperparameter tuning
• Distributed training workflows
• Model compression and optimization

• Fine-tuning LLMs for specialized tasks
• Retrieval-Augmented Generation (RAG)
• Multi-agent AI systems
• Prompt engineering and optimization
• Tool-calling agent architectures
• Memory systems for conversational AI
• Embedding pipelines
• Vector database integrations

• OCR model development
• Document intelligence systems
• Object detection pipelines
• Face detection systems
• AR-based computer vision systems
• Image preprocessing and enhancement

• Model deployment pipelines
• GPU-based training environments
• Scalable inference systems
• API serving for AI models
• AI workflow orchestration
• Self-hosted model deployment

Present

Leading product development, AI architecture, and engineering strategy for scalable AI products and automation systems.

Responsibilities:

• Product strategy & roadmap
• AI product design
• ML pipeline development
• Model deployment architecture
• Workflow automation systems
• Scalable backend engineering
• Team technical direction

Worked on:

• Custom AI model training
• LLM fine-tuning
• OCR intelligence systems
• Automated data pipelines
• NLP systems
• AI agent frameworks
• Predictive automation systems
• Business intelligence models

Built and optimized:

• Real-time virtual try-on systems
• Face landmark detection
• AI-powered fashion overlays
• Computer vision optimization pipelines
• User interaction models

Built production systems for:

• Invoice OCR extraction
• Automated bookkeeping
• Document classification
• Smart financial parsing
• AI-based invoice validation

Currently building LedgerLens, an AI-powered OCR and bookkeeping SaaS.

Built and managed:

• VICIdial infrastructure
• CRM automation systems
• Lead funnel systems
• AI sales bots
• Client onboarding workflows
• Smart communication systems

I actively contribute to open-source technologies and build tools focused on:

• AI infrastructure
• Automation systems
• Backend development
• Developer tooling
• Workflow engines
• Machine learning integrations

🚀 LedgerLens — AI invoice OCR SaaS 🚀 Multi-agent LLM systems 🚀 SmartCredit AI Automation 🚀 AI lead generation pipelines 🚀 Telegram AI bots 🚀 Self-hosted automation infrastructures 🚀 Custom model training systems

Python • PyTorch • TensorFlow • OpenCV • Hugging Face • LangChain • OpenAI APIs • Vector DBs

Node.js • NestJS • FastAPI • Express

Next.js • React • Flutter

PostgreSQL • Prisma • Redis • Firebase

Docker • Linux • Nginx • VPS • GPU Servers

n8n • Zapier • Webhooks • API Integrations

I build intelligent systems that combine research-level AI engineering with real-world business applications — transforming ideas into scalable, production-ready technology.