vishwajeetdabholkar.ai

I build Agentic AI, MCP servers/clients, and real-time RAG systems.

Solutions Engineer passionate about building verifiable AI systems. I create MCP servers/clients, browser automation agents with Playwright, and production-ready RAG systems using real-time data streams. Everything backed by open source code and live demos.

Agentic AI
MCP (Servers/Clients)
Real-time RAG
Web Scraping
Playwright
SingleStore Vectors
Explore projects → Work with me
AGENT://run v0.9 • online

About Me

Solutions Engineer & Data Engineer

I'm a Solutions Engineer at Confluent, specializing in building real-time data streaming and AI solutions. My work focuses on creating intelligent systems that combine Agentic AI, Model Context Protocol (MCP), and Real-time RAG to solve complex problems.

I build production-ready systems using Kafka, Flink, SingleStore, and various AI frameworks. My passion is creating open-source tools and protocols that make AI agents more useful and verifiable. From browser automation with Playwright to vector databases and hybrid search, I focus on shipping working code and demos.

Based in Pune/Kolhapur, India • Available for consulting and collaborative projects

Featured Projects

Across Agentic AI, MCP protocol, RAG, and data engineering
Protocol Engineering

MCP Servers and Clients

Build and customize both MCP servers and MCP clients — so agents really “control your browser”, extract, and query. Integrations for Playwright, Claude, Ollama, and more. See posts for API walkthroughs, protocol guides, and advanced agent demos.

MCPServerClientPlaywright

GitHub → LinkedIn post → Playbooks Playground →

Agentic AI • RAG Tooling

eGet Web Scraper + MCP

Local-first web scraper extracting markdown, JSON, and table content — driven via MCP plug for instant Q&A and RAG ingestion. Browser-based agents for docsites and API data. Demo links show it working with Claude/MCP, Playwright, and Ollama.

FastAPIMarkdownPlaywrightMCP

GitHub → LinkedIn Demo → All LinkedIn activity →

Active RAG / Vectors

Real-time RAG Systems

Real-time retrieval augmented generation: connect semantic + SQL queries, hybrid search, and analytics using SingleStore vectors. See deep dives, blog posts, and live code on high-frequency use cases that go beyond “search + LLM.”

Real-time RAGSingleStoreVectorsHybrid Query

GitHub (PySpark+S3) → LinkedIn RAG Insights → Blog →

AI Utility • Open Source

Resume Parser API

OCR + LLM-driven parser generating structured JSON, ready for ATS and HR flows. Extraction, embedding, and public docs. See LinkedIn posts for API code and integration walkthroughs.

FastAPIOCREmbeddingsOpen Source

LinkedIn Announcement →

Agent Toolkit

Druid Tools for Agents

Agent-safe Druid introspection and SQL execution — sample, stats, and query guardrails inside agent flows for real production workloads.

Apache Druidphidata/AgnoTooling

LinkedIn Post →

Skills & Technologies

Core competencies and tools

AI & ML

Agentic AI LLMs RAG Embeddings Prompt Engineering

Protocols & Agents

MCP Claude Ollama Playwright Browser Automation

Data Engineering

Kafka Flink PySpark Apache Druid Real-time Streaming

Databases & Storage

SingleStore Vector DBs MongoDB AWS S3 Hybrid Search

Development & Tools

Python FastAPI JavaScript TypeScript Git Docker REST APIs

Frameworks & Platforms

phidata/Agno Confluent Cloud AWS Web Scraping OCR

Open Source & Community

Source code, protocol guides, and live agent demos.
Frameworks

phidata → Agno

Hands on with agent frameworks and UI — contributed open source tooling for business-actual agents and live database integrations.

AgnoAgent UITools

Docs →

Protocols

MCP Server Listing

MCP server for web scraping and agent protocols. Compatible with Claude Desktop, Playwright agents, and Ollama API. Explore recent agent demos on my LinkedIn.

MCPClaudeLocal

MCP Listing →

See my latest posts, agent demos, MCP protocol walkthroughs, and code drops on LinkedIn →

Get in touch

Custom agent builds, protocol integrations, consulting.