Agentic Search
Mira's Agentic Search plans its own path across multiple data sources, reasons across a candidate's whole career, and matches by meaning — not keywords. Describe a role in plain language and get a targeted shortlist.
Agentic Search is how Mira finds candidates. You describe a role in plain language; the Agent plans where to look, searches across multiple data sources, and evaluates each candidate by reasoning — not keyword matching.
What makes the search "agentic"
Three things set it apart from a keyword search box:
- Multi-source. The Agent decides which data sources to use for your role and searches across several of them — public talent databases and Mira's own index of public profiles and job postings — instead of querying one fixed database.
- Multi-hop reasoning. It connects evidence across different parts of a candidate's career. "Sales experience at a public company and startup experience" is checked across multiple roles, not matched as a single keyword.
- Reasoning-based matching. Mira Reasoning Embedding (MRE) interprets what you mean. "5 years of experience" is computed from the actual career timeline; "0 to 1" is read as early-stage company building — concepts a keyword search can't capture.
Start a search
A search runs inside a Task — an ongoing conversation between you and the Agent.
There are three ways to start one:
- Click New Task in the top-left of the sidebar to open a fresh conversation.
- Type in the input field on the main screen, then click Send Message.
- Use a suggested scenario below the input — click one to pre-fill a realistic request, then edit it before sending.
Describe the role well
The Agent understands natural language — you don't need a formal job description. But more context gives MRE more to reason about. Include the elements that matter:
| Element | Example | Why it helps |
|---|---|---|
| Role / title | "Senior full-stack engineer" | Defines the search scope |
| Key skills | "React, Node.js, TypeScript" | Narrows technical requirements |
| Experience level | "5+ years" or "has managed a team" | Sets seniority expectations |
| Company type | "high-growth startup" or "enterprise" | Filters by background |
| Location | "Berlin or remote" | Geographic targeting |
| Implicit criteria | "has built something from 0 to 1" | MRE reasons about meaning, not just tags |
You don't need to provide all of these. Mira asks clarifying questions about anything important that's missing.
Weak vs. strong descriptions
| Weak | Strong | Why it matters |
|---|---|---|
| "Find me a developer" | "Senior backend engineer, 5+ years Python, has built microservices at scale, preferably from fintech" | More context gives MRE more signals to match against |
| "Marketing person in NYC" | "B2B SaaS marketing manager in NYC, has run product launches, comfortable with data-driven campaigns" | Implicit criteria trigger reasoning-based matching |
| "We need someone good" | "Someone who's taken a product from 0 to 1, ideally at a Series A–B startup" | Mira reads "0 to 1" as early-stage building experience |
The kinds of criteria MRE understands
- Arithmetic — "At least 5 years in backend development." MRE computes duration from the career timeline, not a profile tag.
- Semantic — "Someone with 0-to-1 experience" or "built a team from scratch." MRE interprets the implied meaning.
- Commonsense — "Public company background" or "high-growth startup DNA." MRE infers context from work history (company size, stage, growth).
- Multi-source — "Sales experience at a public company and startup experience." MRE reasons across different parts of a career.
After you submit
The Agent will:
- Analyze your input and identify what it understands.
- Ask clarifying questions, as a short form, only when key details are missing or ambiguous.
- Draft an Ideal Candidate Profile that consolidates your requirements, including the must-haves candidates have to meet. You can review and edit it.
- Wait for you to confirm before the search starts.
Refine your results
After reviewing a Shortlist, adjust in the same conversation — the Agent keeps the full context and applies your feedback to the next search:
- "Show me more junior candidates."
- "Include people from Amsterdam too."
- "Focus on candidates from e-commerce companies."
- "Narrow it down to people with more startup experience."
Start broad, then tighten. One role per Task keeps each conversation focused and easy to revisit.
Related
- Review your Shortlist — read and export the candidates you get back.
- AI candidate matching — how MRE evaluates fit.
- Agent Teams — put several specialized agents on a complex search.
Ready to try Mira? Join the waitlist for early access.