---

😊 Who I Am

Bridging the gap between AI agent engineering, GenAI infrastructure, and enterprise automation.

🧑‍💻 Role: AI Agent Engineer & AI Architect
🏢 Current: Samsung R&D
💼 Past Experiences: Infosys · Zoho
🧠 Tech Focus: LangChain, Local model hosting, Docker, Kubernetes, GenAI infrastructure, Langfuse, OpenTelemetry, LLM/SLM serving
🤖 Agent Direction: Creating personalized AI agents for every employee, improving agent memory optimization, and building autonomous systems with human control, enterprise safety, and governance
📚 Currently Learning: Model fine-tuning, edge-device model hosting, and optimized local inference
🎯 Current Mission: Building internal AI agents that improve employee productivity and automate enterprise workflows 🚀

---

🚀 Current Work: AI Agents for Enterprise Productivity

At Samsung R&D, I’m working around AI agent creation for internal employee workflows.

Key focus areas:

• Designing AI agent architectures for planning, reasoning, and tool execution
• Building agent workflows using LangChain and Local/private model integrations
• Creating personalized AI agents for employees based on their workflows, context, and productivity needs
• Improving agent memory optimization for better continuity, personalization, and task understanding
• Building autonomous agent systems with full human control, enterprise safety, and workflow governance
• Working across GenAI infrastructure, LLM/SLM serving, Docker, Kubernetes, Langfuse tracing, and OpenTelemetry-based observability
• Learning and exploring Model fine-tuning, edge-device model hosting, and optimized local inference

---

🌟 Selected Focus Areas & Showcase

🚧 Portfolio projects are being organized. Current focus is on internal AI-agent systems, GenAI infrastructure, enterprise automation, observability, and AI architecture workflows. 🚧

🤖 Agent Engineering Tool-using AI agents Planning and reasoning workflows LangChain / LangGraph orchestration Agent memory and context handling Employee productivity assistants Personalized agents for employee workflows Sub-agent and skills-based system design Supervisor-agent orchestration patterns

🏗️ GenAI Infrastructure LLM / SLM serving Local/private model integration vLLM-based inference direction RAG and knowledge workflows Scalable enterprise AI systems Model fine-tuning Local model hosting Edge-device model hosting direction

📊 Observability & Tracing Langfuse-based tracing OpenTelemetry integration Model monitoring Hallucination monitoring Accuracy monitoring RAG output monitoring Retriever performance monitoring Prompt and response quality tracking

🧠 AI Architecture & Enterprise Design Using WebSocket-based communication to reduce latency and improve user experience Working on end-to-end encryption, E2EE, with KMS-based key management systems Designing long-term and short-term agent memory systems -Building and designing sub-agent and skill-based architectures under supervisor-agent control

---

🚀 Tech Stack & AI Toolkit

🤖 AI Agents, LLMs & GenAI

🏗️ GenAI Infrastructure & Local Models

💻 Programming & Engineering

📚 Libraries & Frameworks

---

📊 GitHub Metrics

---

🐍 Contribution Snake

---

⭐ What My GitHub Stars Say About Me

My GitHub stars show a strong interest in building practical, enterprise-ready AI-agent systems with strong architecture, observability, and local/private model support.

🧩 Agent Frameworks & Workflows LangChain / LangGraph workflows Tool-using AI agents Planning and reasoning systems Autonomous workflow engines Supervisor-agent patterns Sub-agent and skills-based design Human-in-the-loop control flows	🛠️ Tooling & Automation MCP servers and tool integrations Browser automation Desktop automation Document automation API-driven workflow automation AI coding workflows Enterprise productivity tooling
🧠 Local, Private & Personalized AI Local LLM / SLM applications Self-hosted AI assistants Private model workflows Local model hosting Edge-device model hosting direction Personalized employee agents Agent memory optimization	📊 Enterprise AI & Knowledge Systems Text-to-SQL systems RAG and knowledge workflows Retriever monitoring RAG output quality monitoring Knowledge assistants Employee productivity agents Workflow intelligence
📈 Observability & Model Quality Langfuse-based tracing OpenTelemetry integration Model monitoring Hallucination monitoring Accuracy monitoring Prompt and response quality tracking Agent execution trace analysis	🏗️ GenAI Infrastructure & Architecture LLM / SLM serving vLLM-based inference direction Docker and Kubernetes deployment Model fine-tuning WebSocket-based low-latency systems E2EE with KMS-based key management Scalable enterprise GenAI architecture

---

💼 Experience

🏢 Samsung R&D

Currently working on AI Agent Engineering and GenAI Infrastructure for internal enterprise productivity.

• Building personalized internal AI agents for employee workflows
• Designing LangChain / LangGraph-based agent systems
• Working with local/private model hosting and enterprise GenAI infrastructure
• Improving agent memory systems for better context, personalization, and continuity
• Designing sub-agent, skills-based, and supervisor-agent orchestration patterns
• Building autonomous agent workflows with human control, safety, and governance
• Working with observability using Langfuse tracing and OpenTelemetry
• Architecting practical GenAI systems for real workplace usage

🏢 Infosys

Previously worked on enterprise GenAI, MLOps, data science, and cloud-based ML systems for large-scale retail use cases.

• Built GenAI-powered Q&A systems using Gemini, LangChain, ChromaDB, FAISS, and RAG-based retrieval workflows
• Worked on internal chatbot systems with document chunking, embeddings, vector search, cross-encoder ranking, and query expansion
• Developed ML solutions for customer behavior prediction, purchase intent classification, and transaction forecasting using large-scale BigQuery datasets
• Designed automated ML workflows using Vertex AI, Kubeflow Pipelines, Docker, Kubernetes, GKE, and CI/CD practices
• Worked on model retraining pipelines, model evaluation, data preprocessing, feature engineering, and scalable ML deployment
• Applied Python, SQL, Pandas, NumPy, Scikit-learn, PyTorch, Flask, BigQuery, and Google Cloud services for data-driven business solutions

🏢 Zoho

Previously worked in a product-focused engineering environment.

• Product engineering
• Software development
• System design foundations
• Practical engineering for real users

---

🧭 Current Learning Roadmap

AI Agent Architecture
LangChain / LangGraph Workflows
Local LLM and SLM Integration
Local Model Hosting
Model Fine-Tuning
Edge-Device Model Hosting
Docker and Kubernetes for GenAI Systems
Langfuse Tracing
OpenTelemetry Observability
RAG Output Monitoring
Retriever Performance Monitoring
Hallucination and Accuracy Monitoring
MCP Tooling
vLLM Model Serving
Enterprise GenAI Infrastructure
Private AI Systems
Personalized Employee Agents
Agent Memory Optimization
Supervisor-Agent and Sub-Agent Design

---

🧠 Engineering Philosophy

Build AI systems that are useful, private when needed, observable, secure, tool-aware, reliable, and practical for real users

I believe the next wave of enterprise AI will move beyond simple chatbots.

It will be powered by agentic systems that can understand context, use tools, remember user preferences, automate workflows, and support employees in real work.

---

📈 Benchmarks, Progress & Contacts

📬 Connect With Me

---

Building AI Agents · Architecting GenAI Systems · Automating Enterprise Workflows