> ## Documentation Index > Fetch the complete documentation index at: https://docs.nosocial.ai/llms.txt > Use this file to discover all available pages before exploring further. # Core Concepts > The fundamental principles behind NoSocial's matching engine. ## Context & Intent CVs and JDs capture history and wishlists. They rarely capture what either side is actually trying to achieve *right now*. NoSocial puts context and intent at the center of matching. **For Candidates:** * What you want next (e.g. “first PM role”, “move from agency to product company”) * Constraints (visa, relocation, caregiving responsibilities) * Workstyle and culture preferences **For Employers:** * Why this role exists (new product, replacement, scaling a team) * Team maturity and expectations * Strategic goals (e.g. “ship X in 6 months”) Most of this information stays private to your AI agent. It is used to shape which matches you see and how they are prioritized. ## Two-Sided Evaluation Most hiring tools are optimized for one side. NoSocial explicitly evaluates every potential match from both perspectives. * **Candidate Score:** How well does this job fit your requirements and constraints? * **Employer Score:** How well does this candidate fit the role and team needs? Matches are only surfaced when there is a strong fit on *both* sides, reducing noise and wasted time. This matching process is slower than a typical "show me jobs or candidates" approach that is a search query. NoSocial performs matching across the entire market and this takees longer. The advantage of this approach is that when you see a match, it is a 2-sided fit that has a 70X lower signal-to-noise ratio vs other platforms and orders of magnitute higher probabilyt for securing an interview. ## Continuous Matching Traditional platforms run a search only when you hit “apply” or “search”. NoSocial runs matching continuously in the background. * **Always On:** As new roles or candidates enter the system, they are immediately evaluated against your criteria. * **Dynamic:** When you update your preferences, your existing matches are re-evaluated. * **Evolving:** Feedback you give on one match improves the next set of matches. ```mermaid theme={null} timeline title Match Quality Evolution Initial Setup : Basic profile & instructions : First matches generated Week 1-2 : Provide feedback on matches : AI learns preferences Week 3-4 : Match quality improves : More relevant opportunities Ongoing : Continuous refinement ``` ## Feedback Driven NoSocial learns from the feedback you provide. For every match you reject or accept, the system refines its understanding of your preferences. * **Refine Understanding:** Discover unstated preferences through patterns. * **Identify Deal-Breakers:** Learn what you definitely don't want. * **Provide Insights:** The AI may share insights like "You tend to prefer smaller companies" based on your behavior. ## Understanding Match Scores 90-100% compatibility. Strong fit on both sides. 70-89% compatibility. Meets most core requirements. 50-69% compatibility. Worth reviewing but may have gaps.