The Missing Protocol: How MCP Bridges LLMs and Data Streams Because Your AI Agents Shouldn’t Be Blind to What’s Happening Right Now Viktor Gamov (@gamussa)

Who’s here? —————————————————————————————————————————— • • • • How many of you are building AI agents right now? (raise your hand) Keep them up - how many are using MCP to connect those agents to external systems? How many of you have streaming data - Kafka, Flink, event-driven architectures? (anyone? okay, good) How many have tried to make an agent react to real-time events? (this is the talk for you) X/Bluesky: @gamussa AgentCon Toronto, 2026 2 / 41

The Hype —————————————————————————————————————————— Everyone is talking about MCP. I have no idea why. (just kidding, I totally know why) Let’s look at the data… X/Bluesky: @gamussa AgentCon Toronto, 2026 3 / 41

Google Trends: “Model Context Protocol” —————————————————————————————————————————— Something happened in November 2024. Anthropic published a spec. And the world lost its mind. X/Bluesky: @gamussa AgentCon Toronto, 2026 4 / 41

Who Am I? —————————————————————————————————————————— Viktor Gamov Developer Advocate, Java Champion, book author I’ve been streaming data since before streaming was cool. (and definitely before it was on the LinkedIn influencer curriculum) X/Bluesky: @gamussa AgentCon Toronto, 2026 5 / 41

Slides and Resources —————————————————————————————————————————— Everything from this talk - slides, links, resources speaking.gamov.io X/Bluesky: @gamussa AgentCon Toronto, 2026 6 / 41

Part 1: The Problem —————————————————————————————————————————— X/Bluesky: @gamussa AgentCon Toronto, 2026 7 / 41

Your Agents Are Impressive. And Blind. —————————————————————————————————————————— Your agents can: • • • • Query databases (stale by the time you ask) Search the web (everyone else’s data) Read your docs (written last quarter) Call APIs (request/response, point-in-time) But they cannot see what’s happening in your systems right now. X/Bluesky: @gamussa AgentCon Toronto, 2026 8 / 41

The Evolution (and the Gap) —————————————————————————————————————————— Every layer solves one problem. But the API is still stateless. Each call starts fresh. RAG gives context, agents give actions - MCP gives a standard way to connect both. X/Bluesky: @gamussa AgentCon Toronto, 2026 9 / 41

Why Streaming Data? —————————————————————————————————————————— Your business doesn’t run in batch. Events happen continuously. What’s streaming? • • • • Where does it live? Orders placed Sensors reporting Users clicking Payments processing • • • • Apache Kafka (most common) Apache Flink (processing) Schema Registry (contracts) Topics, not tables This is the data your agents are missing. X/Bluesky: @gamussa AgentCon Toronto, 2026 10 / 41

The Integration Nightmare —————————————————————————————————————————— Every AI tool vendor says: “Just use our SDK!” So you write custom code for: • • • • • Claude API OpenAI API Gemini API Llama via Ollama Mistral API • • • • • Your Your Your Your Your vector DB Kafka cluster Flink jobs monitoring internal APIs That’s N × M integrations. Does it make sense? No. It’s a nightmare. X/Bluesky: @gamussa AgentCon Toronto, 2026 11 / 41

Remember USB-A? —————————————————————————————————————————— Have you ever tried to put a USB-A cable in correctly on the first try? I’ll give you a moment. You turn it one way. Doesn’t fit. You turn it the other way. Still doesn’t fit. You turn it back. Now it fits. Thank you, European Union. We finally have USB-C. MCP is USB-C for AI integrations. X/Bluesky: @gamussa AgentCon Toronto, 2026 12 / 41

Part 2: Enter MCP —————————————————————————————————————————— X/Bluesky: @gamussa AgentCon Toronto, 2026 13 / 41

What is MCP? —————————————————————————————————————————— Model Context Protocol - an open standard by Anthropic • • • • • A protocol, not a product JSON-RPC based (I know, I know… JSON-RPC in 2026… but hear me out) Client-Server architecture Language agnostic One protocol to connect LLMs to everything X/Bluesky: @gamussa AgentCon Toronto, 2026 14 / 41

MCP Architecture —————————————————————————————————————————— X/Bluesky: @gamussa AgentCon Toronto, 2026 15 / 41

The Three Primitives —————————————————————————————————————————— ▒▒▒▒ Tools ▒▒▒▒ Resources ▒▒▒▒ Prompts Functions the LLM can call Data the LLM can read Templates the LLM can use × query_topic() × send_message() × create_pipeline() X/Bluesky: @gamussa × topic://orders × schema://avro/user × metrics://lag AgentCon Toronto, 2026 × debug_consumer × analyze_throughput × explain_schema 16 / 41

Part 3: MCP Meets Data Streaming —————————————————————————————————————————— X/Bluesky: @gamussa AgentCon Toronto, 2026 17 / 41

What If… —————————————————————————————————————————— What if your AI agent could: • • • • • Subscribe to a Kafka topic and react to events Query a Flink materialized view in real time Understand your Avro schemas without you explaining them Monitor consumer lag and alert you before things break Create new streaming pipelines from natural language X/Bluesky: @gamussa AgentCon Toronto, 2026 18 / 41

MCP + Kafka Architecture —————————————————————————————————————————— X/Bluesky: @gamussa AgentCon Toronto, 2026 19 / 41

Real Example: Anomaly Detection —————————————————————————————————————————— You: “Hey Claude, anything weird happening in the orders topic?” Claude (via MCP): ▍ ▍ ▍ ▍ ▍ ▍ ▍ ▍ I checked the orders topic. In the last 5 minutes: * Order volume dropped 73% compared to the hourly average * Consumer group payment-processor has lag of 45,000 messages * Last successful commit was 8 minutes ago This looks like the payment service is down. Want me to check the payment-service-health topic? All from live streaming data. Not a dashboard. Not a runbook. An agent. X/Bluesky: @gamussa AgentCon Toronto, 2026 20 / 41

Part 4: Let’s Build It —————————————————————————————————————————— X/Bluesky: @gamussa AgentCon Toronto, 2026 21 / 41

Kafka MCP Server Config —————————————————————————————————————————— { “mcpServers”: { “confluent-kafka”: { “command”: “npx”, “args”: [ “-y”, “@confluentinc/mcp-confluent”, “-e”, “.env” ] } } } X/Bluesky: @gamussa AgentCon Toronto, 2026 22 / 41

Two Flavors, Same API —————————————————————————————————————————— ▒▒▒▒ Confluent Cloud ▒▒▒▒ OSS / Self-managed @confluentinc/mcp-confluent mcp-kafka • • • Managed Kafka, Flink, Schema Registry Confluent Cloud authentication Full platform integration • • • Any Kafka cluster Open source Same tools, same API surface github.com/gamussa/mcp-kafka github.com/confluentinc/mcp-confluent They’re tool-compatible - same MCP interface, swap the server, keep your agent code. X/Bluesky: @gamussa AgentCon Toronto, 2026 23 / 41

What Tools Does It Expose? —————————————————————————————————————————— × × × × × × × × list-topics consume-messages produce-message list-schemas create-flink-statement get-consumer-group-lag create-connector diagnostics → → → → → → → → See all topics in the cluster Read from any topic Write to any topic Browse Schema Registry Run Flink SQL Monitor consumer health Deploy connectors Troubleshoot errors The LLM discovers these tools. You don’t hardcode anything. It’s kind of like giving your AI a backstage pass to your data platform. X/Bluesky: @gamussa AgentCon Toronto, 2026 24 / 41

Flink SQL via MCP —————————————————————————————————————————— Ask your agent: “Create a real-time dashboard for flight delays” The agent writes and submits: INSERT INTO flight_dashboard SELECT airline_name, CASE WHEN dep_delay > 0 THEN ‘DELAYED’ ELSE ‘ON_TIME’ END AS status, CASE WHEN HOUR(dep_time) < 12 THEN ‘morning’ WHEN HOUR(dep_time) < 17 THEN ‘afternoon’ ELSE ‘evening’ END AS time_of_day, COUNT(*) AS flight_count FROM flights GROUP BY airline_name, status, time_of_day; X/Bluesky: @gamussa AgentCon Toronto, 2026 25 / 41

You Favourite OLAP Pipeline —————————————————————————————————————————— Kafka + Flink + Pinot (or your favorite OLAP) MCP doesn’t replace any of these. It orchestrates them. X/Bluesky: @gamussa AgentCon Toronto, 2026 26 / 41

Demo: The Brewmaster Agent —————————————————————————————————————————— X/Bluesky: @gamussa AgentCon Toronto, 2026 27 / 41

Meet the Brewmaster —————————————————————————————————————————— An AI agent that manages a craft brewery’s streaming platform. The setup • • • • The agent Confluent Cloud cluster Topics for orders, inventory, sensors Flink SQL for real-time analytics Schema Registry for Avro schemas • • • • LangChain4J + TUI4J + Antropic APIs @confluentinc/mcp-confluent Natural language only No code, no dashboards Let’s see what happens when you talk to your data platform like a human. X/Bluesky: @gamussa AgentCon Toronto, 2026 28 / 41

Demo: Exploring the Cluster —————————————————————————————————————————— “Hey, what topics do I have in my brewery cluster?” (live demo - agent discovers topics via list-topics) “What does the schema for the brewery-sensors topic look like?” (live demo - agent reads the schema via list-schemas) X/Bluesky: @gamussa AgentCon Toronto, 2026 29 / 41

Demo: The Brewmaster in Action —————————————————————————————————————————— “A batch of IPA just finished fermenting. Produce a message to the inventory-updates topic.” (live demo - Claude composes the message, matches the schema, produces via produce-message) “Create a Flink SQL job that monitors for temperature anomalies — alert if any fermenter drifts more than 2 degrees from its style’s normal range in a 10-second window.” (live demo - Claude writes and submits Flink SQL via create-flink-statement) X/Bluesky: @gamussa AgentCon Toronto, 2026 30 / 41

Demo: When Things Go Wrong —————————————————————————————————————————— “Something feels off with the temperature sensors. Can you check?” (live demo - agent consumes from sensor-readings, spots anomalies, checks consumer lag) “Create an alert pipeline that flags readings above 75 degrees.” (live demo - agent creates a Flink SQL statement for the alert) X/Bluesky: @gamussa AgentCon Toronto, 2026 31 / 41

Part 5: Honest Trade-offs —————————————————————————————————————————— X/Bluesky: @gamussa AgentCon Toronto, 2026 32 / 41

What Works —————————————————————————————————————————— • • • • • ✓ ✓ ✓ ✓ ✓ Schema-aware queries - agent understands Avro/Protobuf automatically Ad-hoc exploration - “show me the last 10 messages from topic X” Pipeline creation - natural language to Flink SQL Cross-system correlation - “why is consumer lag growing?” Debugging - agent reads logs, schemas, configs in context X/Bluesky: @gamussa AgentCon Toronto, 2026 33 / 41

What Doesn’t (Yet) —————————————————————————————————————————— • • • • ✗ High-throughput consumption - MCP is request/response, not streaming ✗ Latency guarantees - LLM reasoning adds seconds, not milliseconds ✗ Autonomous production writes - do you really want an LLM producing to your orders topic? ✗ Complex stateful processing - Flink is better at this than any agent MCP is for orchestration and exploration, not replacing your streaming pipeline. X/Bluesky: @gamussa AgentCon Toronto, 2026 34 / 41

When to Use What —————————————————————————————————————————— Use Case │ Tool ───────────────────────────────┼───────────────────── Real-time event processing │ Flink Persistent streaming pipelines │ Kafka + Flink Ad-hoc data exploration │ MCP + LLM Pipeline creation & debugging │ MCP + LLM Anomaly investigation │ MCP + LLM Automated alerting │ Flink (not an agent) “What happened at 3am?” │ MCP + LLM The right tool for the right job. Not everything needs AI. But some things really do. X/Bluesky: @gamussa AgentCon Toronto, 2026 35 / 41

Why This Matters —————————————————————————————————————————— We’re at an inflection point. Before MCP - AI and streaming are separate worlds. Your dashboards show data. Your agents answer questions. They don’t talk to each other. After MCP - AI agents become first-class citizens of your data platform. They can read streams, understand schemas, create pipelines, and investigate incidents. The streaming platform becomes the nervous system. MCP becomes the interface between human intent and machine execution. X/Bluesky: @gamussa AgentCon Toronto, 2026 36 / 41

What to Do Monday Morning —————————————————————————————————————————— 1. 2. 3. 4. 5. Install the Confluent MCP server - npx @confluentinc/mcp-confluent Connect Claude Desktop to your dev cluster (not prod… yet) Ask it questions - “What topics do I have? What’s the schema for X?” Build an agent for your own streaming use case Come talk to me - I’ll be around after the talk X/Bluesky: @gamussa AgentCon Toronto, 2026 37 / 41

Resources —————————————————————————————————————————— • • • • • MCP Spec - modelcontextprotocol.io Confluent MCP github.com/confluentinc/mcp-confluent OSS Kafka MCP - github.com/gamussa/mcp-kafka Confluent Cloud - confluent.io/confluent-cloud This talk - speaking.gamov.io X/Bluesky: @gamussa AgentCon Toronto, 2026 38 / 41

As always, have a nice day. —————————————————————————————————————————— Viktor Gamov - X/Bluesky: @gamussa

Questions? —————————————————————————————————————————— I’ll be around after the talk. Find me on X/Bluesky: @gamussa

X/Bluesky: @gamussa AgentCon Toronto, 2026 41 / 41