Manoj Karki
Senior Web Scraping & Data Pipeline Engineer
I engineer high-throughput web crawlers, custom Apify Actors, and Model Context Protocol (MCP) data connectors streaming live web intelligence directly into vector databases, AI agents, and local Hermes execution models.
Web Scraping, AI Agents & Data Engineering
Specialized technical skillset spanning large-scale data extraction, custom actor development, anti-bot navigation, and AI agent integration.
Web Scraping at Scale
AI Agents & Local Models
Vector DBs & RAG
Automation & Middleware
Battle-Tested Web Scraping Actors
Production-grade extraction tools built with proxy rotation, dynamic DOM parsing, and cookie-free browser emulation.
Instagram Profile Email & Phone Scraper
Query public GraphQL endpoints to extract public business contact emails, phone numbers, bio categories, and verification status without account ban risks.
Instagram Following Scraper
Stream complete account following lists and social connectivity graphs from public profiles with zero account credentials or cookies needed.
Email Verify & Validate (Bulk)
Execute real-time MX record lookups, SMTP handshake simulations, catch-all server checks, and domain risk scoring to protect sender reputation.
TikTok Profile & Contact Scraper
Bypass dynamic TikTok signature algorithms using headless Playwright sessions to harvest creator contact details, video play metrics, and bio links.
MCP & Local Hermes AI Agent Data Feeds
Stream live structured web payloads directly into AI agent context windows using Anthropic's Model Context Protocol (MCP) or run localized privacy-focused automation with the Hermes Agent Framework.
Claude Desktop & Cursor IDE
Connect scrapers directly into developer chats via standard MCP JSON-RPC protocols, enabling natural language live web data extraction inside IDEs.
Hermes Local Agent Runtimes
Deploy local AI agents powered by Ollama and Qwen models using the Hermes Framework for local offline execution without SaaS API token costs.
RAG & Vector DB Ingestion
Automated text extraction, chunking, and vector embedding distribution streaming into Pinecone, Qdrant, and PostgreSQL pgvector storage.
Start Your Custom Scraper or AI Pipeline
Provide your project details and technical requirements below to receive a direct architecture review, feasibility report, and project quote.