Invenio Search Group: Multi-Agent Recruiting OS With Voice Screening, Agent Hierarchy, and A2A + MCP Architecture

Invenio Search Group ran executive searches on infrastructure most firms left behind in 2010. Applicant tracking happened in spreadsheets. Client relationships lived in individual consultants' heads and email inboxes. Candidate files were literal paper. There was no CRM, no ATS, and zero automation across any stage of the recruiting lifecycle.

The firm had real expertise and a strong client roster. What it didn't have was scalable infrastructure. Every placement required manual coordination. Every new consultant onboarded meant rebuilding tribal knowledge from scratch. Growing the business meant hiring more people to run the same broken manual processes — which is the definition of scaling the wrong thing.

Off-the-shelf ATS products would have solved 30% of the problem. The other 70% — automated AI phone screens, multi-provider candidate sourcing, an agent hierarchy that actually does recruiting work, and a unified inbox-calendar-CRM-ATS — didn't exist as a product Invenio could buy.

Aries Consulting Group deployed a full Custom Platform Build — a multi-agent recruiting OS, not a CRM with AI features bolted on:

• Executive→Specialist agent hierarchy. An Executive Agent orchestrates two senior specialists — a Senior Recruiting Agent (executive search, relationship building) and a Technical Recruiting Agent (technical roles, skill matching) — which delegate to four production specialists: Screening Agent (resume parsing, skill assessment), Communication Agent (email automation, scheduling), Analytics Agent (data analysis, reporting), and Research Agent (market research, sourcing). 15+ agents total, all visible in a Live Agent Hierarchy view with success rates and active job counts per agent.
• Automated voice screening via ElevenLabs + Twilio. The platform places initial phone screens autonomously, transcribes the call, and runs AI analysis that produces a numeric score, a recommendation (Strong / Recommended / Not Recommended), and structured strengths and concerns. Recruiters review screened candidates instead of running first-round phone screens themselves.
• Five AI search providers, model-agnostic. Candidate sourcing runs across Gemini, OpenAI, Lutra, Scrapeless, and a custom job-search agent in parallel — the platform isn't locked to one model or one data provider, and adding a new one is a config change, not a rebuild.
• Inbox + calendar replacement. Full Microsoft 365 and Gmail OAuth integration with email-routing, thread-processing, and calendar-availability agents. Recruiters work out of Invenio instead of switching between Outlook, Gmail, and Calendly.
• AI-prioritized work queue. The dashboard's "AI Suggested Activities" ranks the recruiter's next actions by priority (Prepare offers, Review applications, Follow up with prospects, Update job descriptions) so the team works on the highest-leverage thing every hour.
• Built on A2A (Agent-to-Agent) protocol + MCP. Agents communicate via the open A2A spec; tools and data sources are exposed through MCP servers. The platform is the production proof of the "MCP-ready for the agentic era" claim, not a slide.
• Multi-tenant infrastructure with per-user agent provisioning — each consultant gets their own agent instances, their own provider credentials, their own scheduled model-sync runs.
• Ongoing Growth Retainer to add new agents and modules as the business discovers new leverage points.

• +85% increase in applicant placement rate — more candidates placed per search
• +90% improvement in applicant-to-role fit — AI-assisted matching surfacing candidates human consultants were missing
• +30% growth in new client acquisition — the platform freed consultant time from admin work, which got reinvested in business development
• Paper-to-platform transformation completed in one build — no multi-phase migration, no parallel systems
• 15+ AI agents in production across an Executive→Specialist hierarchy with live success-rate telemetry per agent
• First-round phone screens fully automated via ElevenLabs voice + AI transcript analysis

Three things made this engagement deliver results a generic ATS rollout wouldn't have:

1. Agents, not chatbots. Most "AI recruiting" products add a chat window that summarizes resumes. Invenio has 15+ agents in a hierarchy, each with a defined job, success-rate telemetry, and the authority to take actions on live business data — running phone screens, drafting emails, scoring candidates, monitoring pipeline health.
2. Voice screening removes the worst hour of every recruiter's day. The ElevenLabs + Twilio integration places the initial phone screen autonomously and hands the recruiter a scored, structured analysis. That single capability is the largest single time win in the platform — and it does not exist as a product anyone can buy.
3. Model- and protocol-agnostic by construction. Five AI search providers running in parallel. A2A for agent communication. MCP for tool exposure. When a better model ships next quarter, Invenio plugs it in. When the agentic era turns into the agentic decade, Invenio is already on the rails.

What an Agent Hierarchy Actually Does in Production

For the technical reader: the system is structured as orchestration, not a flat collection of tools.

• Executive Agent — owns strategy and coordination. Decides which senior agent should handle a new search.
• Senior Recruiting Agent — executive search, relationship building. Routes work to specialists.
• Technical Recruiting Agent — technical assessment, skill matching. Routes work to specialists.
• Screening Agent — resume parsing, skill assessment, AI voice screens with scoring.
• Communication Agent — email automation, scheduling, follow-ups, candidate comms.
• Analytics Agent — data analysis, reporting, recruiter performance.
• Research Agent — market research, candidate sourcing across five AI providers.

This is what "AI-native" means in practice: not a feature, not a chat window — a hierarchy of agents doing specific work on live business data, every one of them measurable and replaceable.

Invenio Search Group fits the Aries Custom Platform Build ICP: a founder-led professional services firm doing $2M–$50M in revenue, with operations that depend on specialized knowledge and client relationships, currently running on a patchwork of generic tools (or, in Invenio's case, no tools at all).

If you're running a services business on spreadsheets and email — whether recruiting, consulting, agencies, or professional services — and the operational gap is the ceiling on growth, this is the engagement that removes the ceiling.