Amy by Jellyfish

Product Introduction

1. Definition: Amy by Jellyfish is an AI-powered talent sourcing and recruitment automation platform designed specifically for staffing agencies and recruiting teams. It functions as a full-cycle AI sourcer, operating within the technical categories of Recruiting Automation (RecruiterTech) and Applied AI for Human Resources.
2. Core Value Proposition: It exists to automate the entire operational workflow of a human sourcer, transforming unstructured client requirements into a continuous, evidence-based talent discovery and engagement engine. Its primary value is increasing recruiter capacity and candidate quality by handling time-intensive tasks like multi-platform research, candidate qualification, personalized outreach, and inbox management.

Main Features

1. Multi-Platform Evidence-Based Sourcing: Amy searches beyond traditional resume databases like LinkedIn. It actively crawls and analyzes profiles and activity on GitHub (for technical work and open-source contributions), Google Scholar (for research depth and expertise), technical communities (Discourse, Stack Overflow, etc.), and the open web (portfolios, conference talks). How it works: The AI parses client briefs to identify key skills and experience, then uses this to search for tangible evidence of ability across these platforms, ranking candidates based on match quality.
2. Automated Candidate Qualification & Evidence Matching: The platform doesn't just list profiles; it evaluates and explains each candidate's fit. For every potential match, Amy generates an "evidence-backed" summary, listing specific, verifiable accomplishments (e.g., "Built scalable data pipelines with Python and AWS") and mapping them directly to the client's stated requirements. This translates subjective job descriptions into objective, reviewable data points for recruiters and hiring managers.
3. Intelligent Outreach & Candidate Inbox Management: Amy automates the candidate engagement sequence. It personalizes outreach messages (email, LinkedIn Connection, InMail) by incorporating the specific evidence points found during sourcing. It then manages the candidate's replies, answering routine questions, scheduling follow-ups, and only escalating highly-qualified, interested candidates to the human recruiter for a direct conversation. This creates a "handoff" only when a candidate is warm and qualified.

Problems Solved

1. Pain Point: Staffing agencies face inefficient, manual sourcing processes that rely heavily on expensive human sourcers searching the same saturated databases (e.g., LinkedIn Recruiter), leading to high cost-per-hire, slow time-to-fill, and missed "hidden" talent.
2. Target Audience: The primary user personas are Recruiting Agency Owners, Staffing Managers, and Technical Recruiters at staffing and recruiting firms who need to scale their operations without linearly increasing headcount. It is also built for Sourcers within those agencies to augment their capabilities and focus on high-judgment tasks.
3. Use Cases: Essential for agencies filling hard-to-source technical roles (e.g., Senior Data Engineers, Machine Learning Researchers, DevOps Architects) where evidence of skill is found in code repositories and research papers. It is also critical for managing high-volume requisitions or building continuous talent pipelines in niche markets, operating across different time zones without manual intervention.

Unique Advantages

1. Differentiation: Unlike standalone candidate databases (e.g., LinkedIn Recruiter, SeekOut) or simple AI resume screeners, Amy performs the full operational job of a sourcer. Competitors provide a list of profiles; Amy provides a list of explained, evidence-backed, pre-engaged candidates. It moves from a "search tool" model to an "autonomous agent" model for recruitment.
2. Key Innovation: Its core innovation is the closed-loop, context-aware workflow. The original client brief and all subsequent feedback (hiring manager decisions, recruiter notes) remain attached to the search, allowing the AI to learn and refine its matching algorithm in real-time. This creates a living search strategy that improves with use, rather than treating each search as a discrete, forgotten event.

Frequently Asked Questions (FAQ)

1. How does Amy by Jellyfish source candidates beyond LinkedIn? Amy uses AI to search for tangible evidence of skills across multiple platforms, including GitHub for code and project history, Google Scholar for publications and research, technical community forums for peer recognition, and the broader open web for portfolios and professional content, providing a more holistic view of a candidate's abilities than a resume alone.
2. What is the difference between an AI sourcer like Amy and a traditional candidate database? A traditional database is a passive tool requiring recruiters to manually craft boolean searches, review hundreds of profiles, and initiate all outreach. An AI sourcer like Amy is an active agent: it interprets the job requirement, autonomously discovers and qualifies candidates using evidence, personalizes and sends outreach, and manages the initial reply conversation, only handing off warm leads.
3. Can recruiters maintain control and review the AI's work? Yes, Amy is designed for controlled autonomy. Agencies can configure workflows to require human approval at any stage—such as reviewing candidate matches before outreach, editing personalized messages, or approving candidates before they are sent to the client—allowing teams to balance automation with human oversight.
4. What kind of input does Amy need to start a search? To initiate a candidate search, Amy requires a client brief. This can be a formal job description, notes from an intake call, or even a bulleted list of must-have skills, nice-to-have experience, and key role context (e.g., team structure, project focus). The more nuanced the information, the more precise the AI's sourcing becomes.
5. Is Amy suitable for non-technical roles? While its ability to parse GitHub and Google Scholar is optimized for technical and research-focused roles, its underlying architecture for parsing requirements, searching the web, and managing engagement sequences can be applied to other professional domains where online evidence of skill exists, such as design (Behance, Dribbble) or marketing (public blogs, campaign portfolios).