LEAD GENERATION
Agent-built research brief for ML paper authors
On request, a Paperclip agent researches a named ML lab or topic, harvests its prolific paper authors via Brave, cross-checks each against HuggingFace activity.
How it runs
The automated pipeline, trigger to output.
- TriggerOperator requests research on a lab/topic
- ActionAgent runs iterative Brave author researchBrave Search
- ActionConfirm active builders via HuggingFaceHugging Face
- LogicAgent qualifies and drafts outreach brief
- OutputFile qualified leads + briefs in AttioAttio
What it does
This is an agent-driven sourcing job: given a lab name or research topic, the CEO agent uses Brave to find the lab's most-published authors, confirms who is actively shipping by checking their HuggingFace profile, writes a short personalized outreach brief for each, and saves the qualified people into Attio.
When to use it
Choose this over the deterministic harvesters when you want judgment, not just rows — a researched narrative on each person, a reason-to-reach-out, and a quality bar that a fixed pipeline cannot apply on its own.
How it works
- 1The operator asks the agent to research a specific lab or topic.
- 2The agent runs iterative Brave searches to build the author roster and rank by publication volume and recency.
- 3For each candidate it checks HuggingFace for active models, datasets, or Spaces to confirm they still build.
- 4The agent decides who clears the bar and drafts a tailored one-paragraph outreach brief per person.
- 5Qualified leads, with brief and evidence links, are written to Attio.
- 6The output is a curated, talking-points-ready lead set rather than a raw list.
Set it up
What you configure once, before turning it on.
- 1Connect Brave SearchWeb, news, image, video search.
- 2Connect Hugging FaceModels, datasets, spaces — the open-source hub.
- 3Connect AttioReal-time CRM with structured data + powerful views.
- 4Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
- 5Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
- 6Test, then turn it onRun once against a sample, confirm the output, then enable the trigger.
More Lead Generation workflows
Webhook-triggered Brave rising-keyword check into a Notion trend queue
When an external trend or alert tool fires a webhook with a keyword, checks Brave for current intent volume and freshness, has an LLM judge whether it's a real warm signal.
Fuzzy-match badge companies to Salesforce accounts and enrich
Resolves messy hand-typed company names from badge scans to canonical Salesforce accounts using domain and fuzzy-name matching, enriches missing firmographics.
Manual Brave keyword sweep into an Airtable research board
On demand, sweeps a topic across Brave Search, clusters the results by buying stage with an LLM, and writes a deduplicated research board to Airtable with company, source URL.
Daily rollup of scored webinar leads from Airtable into HubSpot lists
On a schedule, read newly scored webinar leads from Airtable, sync each into the matching HubSpot tiered list (Hot/Warm/Cold).
Fast-track hot webinar leads into HubSpot and ping the rep on Slack
After a webinar, identify attendees whose poll answers signal high purchase intent, create or update their HubSpot contact with a lead score.
Classify open-text webinar poll answers with AI and enrich the lead record
For webinars that use free-text poll questions, an AI step reads each attendee's written answers, classifies intent and pain points into structured fields.
Run it inside a business
This workflow drops into a full company template. Import the org, and this is one of the playbooks its agents run.
Run this workflow in your colony.
14-day trial. No DevOps. No Sales call. Provisioned in under a minute.