AI has made many facets of the job search faster, but it hasn't made it easier. You can polish a resume in seconds. Recruiters can screen hundreds of applications in minutes. Everyone's moving faster, and somehow nobody's happy with the state of things.
Here's the problem: so far AI has been applied to all the wrong parts of the hiring process. It's optimizing tasks—writing cover letters, parsing keywords, screening resumes—without solving the actual challenge of connecting the right people to the right opportunities.
We built something that will help you take control.
Dice just launched a Model Context Protocol (MCP) server that connects your AI assistant—Claude, ChatGPT, Gemini—directly to our tech-only job database. You already use AI to write code, debug problems, and automate workflows. Now you can use it to search for jobs the same way: describe what you want in plain language and let your AI do the work.
Click here to see the technical documentation.
Summary
What This Means for Your Job Search
The Dice MCP Server gives AI assistants like Claude and ChatGPT programmatic access to Dice's curated database of technology jobs. Instead of guessing which keywords will surface the right roles, you can search the way you think.
Want to find senior backend roles in Denver that don't require on-call rotations? Looking for remote Python positions at companies willing to sponsor visas? Need to exclude jobs that require Java because you're done with Spring Boot forever?
Just ask.
The MCP server supports natural language queries and gives you access to 13+ filter parameters including location radius, workplace type (remote/hybrid/on-site), employment type, visa sponsorship, and posting recency. You can build searches as simple or as sophisticated as you need.
Here's what that looks like in practice:
Instead of this: Manually checking Dice every morning, trying different keyword combinations, copying job descriptions into spreadsheets, and hoping you didn't miss anything.
You can do this: Open your AI tool and put in a prompt like the following "Check Dice for new senior DevOps roles posted in the last 3 days within 25 miles of Austin, Texas. Focus on companies offering hybrid work arrangements."
The AI handles the query construction, retrieves the results, and can even help you compare opportunities or identify patterns across listings.
Why MCP Changes How You Search for Tech Jobs
Model Context Protocol is an open standard that allows AI assistants to securely connect to external data sources. Think of it as a universal adapter that lets your AI talk to different services without you having to build custom integrations for each one.
For job search, this is a huge deal. You're no longer limited to what a website's search bar can understand. Your AI assistant can:
Translate your conversational requests into precise database queries
Apply multiple filters simultaneously without making you click through dropdown menus
Return structured data that you can ask follow-up questions about
Remember your preferences across searches
Automate recurring searches so you're always aware of new opportunities
It can also:
Tell you how your resume stacks up against the jobs you want.
Recommend upskilling opportunities based on real-time job postings.
Help you conduct a search based on things you don’t want in a role.
The barrier to entry is low. If you're already using Claude Desktop, Claude Code, or a compatible AI assistant, you can connect to the Dice MCP server and start searching in minutes.
From Keyword Matching to Intelligent Search
Traditional job search engines rely on keyword matching. They look for exact words or close synonyms in job titles and descriptions, then rank results based on how many times those keywords appear. This approach has obvious limitations:
You have to guess which exact terms employers used in their job postings
You miss relevant opportunities because you didn't think to search for a specific keyword
You can't easily filter by nuanced preferences like "startups with fewer than 100 employees" or "companies that don't require whiteboard interviews"
You waste time manually reviewing listings that match your keywords but fail other important criteria
Agentic AI search works differently. Instead of matching keywords, the AI understands the intent behind your request and constructs queries that capture what you actually mean. It can reason about trade-offs, apply complex filtering logic, and surface opportunities that keyword search would miss.
This matters especially for tech professionals who work across multiple domains or have preferences that don't map neatly to job titles. If you're a backend engineer who's open to infrastructure roles but wants to avoid anything customer-facing, good luck expressing that as a keyword string. With an AI assistant connected to Dice's MCP server, you just explain what you want.
What You Can Build
The real power of connecting AI to job search infrastructure is automation. Once your AI assistant can access Dice's database, you can build workflows that run continuously without your intervention:
Morning job digests: Get a summary of new postings that match your criteria, delivered before you've had your first coffee
Market monitoring: Track hiring trends in your specialization across different cities or company types
Salary intelligence: Compare compensation ranges for your role across geographic regions
Continuous search: Set up persistent queries that notify you when your ideal role appears, even if you're not actively looking
These aren't theoretical possibilities. They're practical automations you can implement today using the Dice MCP server and readily available AI assistants.
Built for Tech Professionals
Dice has always maintained a tech-only job database. We don't dilute our listings with every type of role under the sun. Every job on Dice is technology-focused, posted by verified employers, and quality-controlled to ensure relevance.
The MCP server maintains that standard. When your AI searches Dice through MCP, it's accessing the same carefully curated database you'd find on dice.com.
This focus matters. You're searching a database built specifically for technology professionals, by people who understand the industry.
Getting Started
The Dice MCP server is live and ready to connect to your AI assistant. Setup takes just a few minutes, and you can start searching immediately.
If you are already comfortable with MCP servers, you can access everything you need in the technical documentation here.
Is this your first time connecting an MCP server? We created a comprehensive guide for connecting to Claude, ChatGPT, and Gemini, complete with setup instructions, example queries, troubleshooting tips, and best practices for getting the most out of agentic search.
The Future of Job Search
Agentic AI changes everything about job search. Instead of making people like you speak the language of search algorithms, we can finally build systems where the AI translates human intent into database queries. You describe what you want and the AI figures out how to find it.
The Dice MCP server is one way we are realizing that future. We're giving tech professionals the tools to take control of their job search—to search the way they think, automate the tedious parts, and focus their energy on the best opportunities.