Customer Success Engineer at IBM. 18+ years across software, data and AI — now designing agent systems that are secure, observable and built to scale.
I'm a technical solution architect with 18+ years across software, data, analytics and AI. My work lives where ambitious ideas meet real systems — defining architectures for large distributed platforms and turning intricate business problems into solutions that ship.
Today my focus is enterprise agentic AI: systems where specialized agents plan, call tools, remember context and coordinate to get real work done. I care most about the parts teams often skip — security, observability and the operational backbone that lets agents run reliably at scale.
I work as a Customer Success Engineer at IBM, closest to the people actually adopting this technology. I like to experiment, then make it production-grade.
From orchestrating fleets of agents to the architecture and cloud foundations underneath them — and the leadership to land it with real teams.
Designing systems of autonomous agents that reason, act and coordinate.
From classical ML to generative models across every modality.
Enterprise-grade foundations for systems that have to hold up.
Connecting technology to the people and outcomes that depend on it.
AI that no longer just answers — it plans, calls tools, remembers, and coordinates with other agents to finish real work. Here's how I approach building it for the enterprise.
Intent arrives
Decompose the goal
Route to agents
Call tools & APIs
Persist context
Trace & evaluate
Coordinating specialized agents — planners, doers, reviewers — into reliable workflows with watsonx Orchestrate and its Agent Development Kit.
Letting agents, tools and systems speak the same language with open standards — MCP for tools, A2A for agent-to-agent, and the wider protocol landscape.
Monitoring, evaluation and telemetry so you can see what agents actually do — the part most AI projects overlook, and the one that decides success at scale.
Treating safety, identity and governance as first-class: agentic systems that are secure, observable and scalable from day one.
Agents are only as useful as what they can reach. Open protocols let them speak to tools, data, and each other without bespoke glue.
Model Context Protocol
Agents ↔ tools, data and APIs through one standard interface.
Agent-to-Agent
Direct, discoverable collaboration between specialized agents.
Agent Communication
Structured, session-aware messaging for multi-turn workflows.
Agent Network
Decentralized, peer-to-peer networks of interoperable agents.
IBM · Istanbul, Türkiye
Working hands-on with enterprise clients to architect and adopt agentic AI — multi-agent orchestration on watsonx Orchestrate, interoperability via open protocols, and the security, observability and operations that make it production-ready.
Technical leadership across software, data & AI
18+ years defining architectures for large distributed systems and enterprise analytics — spanning cloud, data engineering, ML/DL, technical consulting and pre-sales, with a consultative approach focused on real business outcomes.
IBM Entrepreneur Award 2025
IBM
Recognized for innovation and impact with enterprise clients.
Enterprise Agentic AI: Secure, Observable, Scalable
EEMKON · ICC Istanbul
Speaker on building enterprise-scale agentic AI systems.
watsonx & Agentic AI
Data & Analytics Summit · Boğaziçi University
Speaker on AI that decides and acts, not just answers.
Always happy to connect on enterprise agentic AI — architecture, orchestration, protocols, or a conference talk. Drop a note and I'll get back to you.
