Dima PersiianovWe went quiet for a stretch, and skipped the April note to roll it into this one. So this covers two months of work in a single update.
The goal we're building toward is simple to say and hard to do: we want Extruct to be the best deep research engine for GTM. That means answers that come back complete, current, and consistent across your entire list, not pieced together from whatever the open web happened to surface that day.
Two big things shipped.
Extruct Has an MCP Server
Extruct agents already research companies. Now you can run that research from wherever you work. The MCP server brings company lookup, search, lookalikes, Deep Search, and AI Tables into Claude or your editor, in plain language.
Install it once and sign in with your Extruct account. The first time you connect, your browser opens for sign-in.
Connector URL:
https://api.extruct.ai/mcp
Claude Code:
claude mcp add --transport http extruct https://api.extruct.ai/mcp
For Claude desktop and web, add a custom connector under Customize → Connectors with the URL above. Cursor, VS Code, and Codex take the same URL in their MCP settings. Full install guide: docs.extruct.ai/build-with-ai-agents/extruct-mcp.
Then open your agent and ask. Drop a research column, or just ask in chat:
- Is this account scaling sales? How many open roles, where, how fast?
- What's their stack, and do they use anything we integrate with?
- Did they raise in the last 12 months, and what round, amount, and lead investor?
- What partnerships or launches have they announced lately?
Enrichments Got Faster, Deeper, and More Complete
We rebuilt how we research the core sources of GTM technographic data: hiring, tech stack, news, and financing. Faster access, deeper coverage, and answers that hold up across a whole list instead of one lucky company.
What gets dramatically better:
- Hiring - not "a few job posts we found," but the complete active req list, categorized by function and seniority, with hiring velocity. You see where a company is actually investing, not a sample.
- Tech stack - the technologies a company really uses, ranked by signal strength across its web presence and job posts, instead of a guess from the homepage.
- News and buying signals - structured events, not headlines: acquisitions, partnerships, launches, exec hires, expansions, funding. Already typed and dated, so they're filterable and countable.
- Financing - the full round history with amounts, investors, and total raised, not the one funding article that happened to rank.
- Reporting structure - we deeply analyze LinkedIn profiles and job descriptions to reconstruct who reports to whom, and who controls the budget. The org map that matters to your sales motion.
- Teams and initiatives - the teams, technologies, and active initiatives inside your target accounts, so you reach the right people with the right message.
And richer LinkedIn company and people coverage on the same backbone.
And this is the output I vibe-coded after about 20 minutes of chatting with the Extruct MCP.
Why It Matters
Stale data tells you a company uses Marketo. That's a logo on a list. It doesn't tell you anything you can act on.
We'd rather read the actual job descriptions and tell you two teams are hiring for that skill right now. That's the difference between knowing a tool is installed and knowing where a company is putting its money this quarter. That's the research engine we're building, and the MCP is how you put it to work without leaving your tools.
Questions or feedback - we're always happy to hear what you're building with Extruct.
