K M Haroon
Building AI agents and pipelines that hold up in the real world.
- AI/ML Engineering
- Agentic AI
- RAG & LLMOps
- MLOps
- Data Analytics
About
Hi, I'm Haroon, an AI/ML engineer who turns research into systems that ship. I build agentic workflows, RAG pipelines and the MLOps plumbing that keeps models honest in production, and I care as much about the person using it as the model behind it.
Tools I build with
- Python
- PyTorch
- TensorFlow
- LangChain
- pandas
- NumPy
- Power BI
- Tableau
- React
- JavaScript
- Node.js
- Git
- GitHub
- Supabase
- Docker
- MongoDB
Journey
-
CloseFuture · 2 rolesOngoing
May 2026 — Now. AI/ML Engineer, Sep 2026 — Now. Working as an AI/ML Engineer at CloseFuture, building intelligent systems and AI-powered product capabilities. My work spans LLMs, Generative AI, AI Agents, RAG/Agentic RAG, agent orchestration, context engineering, MCP/tool integration, model evaluation, and AI system architecture. I focus on taking emerging AI technologies from research and experimentation to working prototypes and production-oriented solutions, with an emphasis on reliability, efficiency, scalability, and real-world applicability. AI Product & Research Intern, May 2026 — Sep 2026. I explore the frontier of artificial intelligence, evaluating emerging tools, experimenting with cutting-edge models, and identifying where AI can create real, scalable impact. My work sits at the intersection of research and product thinking. I don't just study what's new; I ask what's useful, what's ready, and what's worth building. From stress-testing frameworks to surfacing insights that inform strategy, I bring a hands-on approach to every problem I touch. I'm driven by one core belief: the best AI solutions aren't just technically impressive they're intuitive, human-centered, and built to last. That's the standard I hold my work to.
Working as an AI/ML Engineer at CloseFuture, building intelligent systems and AI-powered product capabilities. My work spans LLMs, Generative AI, AI Agents, RAG/Agentic RAG, agent orchestration, context engineering, MCP/tool integration, model evaluation, and AI system architecture. I focus on taking emerging AI technologies from research and experimentation to working prototypes and production-oriented solutions, with an emphasis on reliability, efficiency, scalability, and real-world applicability.
I explore the frontier of artificial intelligence, evaluating emerging tools, experimenting with cutting-edge models, and identifying where AI can create real, scalable impact. My work sits at the intersection of research and product thinking. I don't just study what's new; I ask what's useful, what's ready, and what's worth building. From stress-testing frameworks to surfacing insights that inform strategy, I bring a hands-on approach to every problem I touch. I'm driven by one core belief: the best AI solutions aren't just technically impressive they're intuitive, human-centered, and built to last. That's the standard I hold my work to.
-
LangChainOngoing
Dec 2025 — Now. Contributing to the development of LangChain, an open-source framework for building applications with LLMs, by implementing new features, fixing bugs, and improving documentation to enhance the developer experience and expand the capabilities of the framework.
Contributing to the development of LangChain, an open-source framework for building applications with LLMs, by implementing new features, fixing bugs, and improving documentation to enhance the developer experience and expand the capabilities of the framework.
-
Freelance AI Developer — IBM Watsonx Orchestrate3 months
Sep 2025 — Nov 2025. Developing an enterprise-level multi-agent AI platform using IBM Watsonx Orchestrate to support personalized student learning and university assistance workflows. The system is designed for scalable, secure agent orchestration and is scheduled for release soon.
Developing an enterprise-level multi-agent AI platform using IBM Watsonx Orchestrate to support personalized student learning and university assistance workflows. The system is designed for scalable, secure agent orchestration and is scheduled for release soon.
-
The Mind Company2 months
Jul 2025 — Aug 2025. Built a machine learning driven analysis to study decline in human cognitive retention and focus, and evaluated how simple game-based interventions support attention improvement and cognitive recovery.
Built a machine learning driven analysis to study decline in human cognitive retention and focus, and evaluated how simple game-based interventions support attention improvement and cognitive recovery.
-
Code from Basics, Chennai4 months
Oct 2024 — Jan 2025. Built an interactive Power BI dashboard to analyze user engagement across major social media platforms, designing KPIs, DAX measures, and trend visualizations to uncover insights on content performance, demographic patterns, and platform reach, demonstrating strong analytical thinking, dashboard design expertise, and the ability to translate raw data into actionable marketing insights.
Built an interactive Power BI dashboard to analyze user engagement across major social media platforms, designing KPIs, DAX measures, and trend visualizations to uncover insights on content performance, demographic patterns, and platform reach, demonstrating strong analytical thinking, dashboard design expertise, and the ability to translate raw data into actionable marketing insights.
Writing
Field notes on AI that holds up in production: agents, orchestration and the stack beneath them.
-
Before You Build an AI Agent
An agent is one layer of a production AI system. Here is the stack underneath it and the engineering that decides whether it survives production.
-
How Agents Communicate
The five multi-agent communication patterns, the four orchestration topologies, and the questions that reveal which one you actually built. Every build ran — the layer that decides whether they hold together is the one the room never tested.
Skillset
Stack
- Python
- Machine Learning Algorithms
- LangChain
- LLM-Based Agents
- LangGraph
- Agentic-AI Orchestration
- TensorFlow
- Agentic AI Systems
- PyTorch
- Multi-Agent Architecture
- Scikit-learn
- Feature Engineering
- LLMs
- AI Workflow Orchestration
- Pandas
- NumPy
- MySQL
- MongoDB
- Supabase
- Power BI
- Tableau
- EDA
- Data Engineering
- Streamlit
- Docker
- GitHub Actions
- REST API
- Data Pipelines
- Model Deployment
- Cloud Deployment
- Model Registry
- Experiment Tracking
Haroon
OPEN THE DOOR
Share your details, and I’ll get back to you personally, within 24 hours.
GOT IT.
Thanks — your details are in. I’ll reply personally within 24 hours.