Three pains every sales and BD team hits weekly. Each one is what your reps actually complain about, not what a feature page would call them.
Recruiters burn six hours a week on sourcing lists and rubric scoring before a single screen call, and Tier-1 candidates ghost while the dossier is still being assembled.
Attrition shows up in the resignation email, not in a 90-day flight-risk panel, so retention budget gets spent on goodbye gifts instead of equity refresh.
Offer letters and onboarding packs ship inconsistent comp bands and missing GDPR clauses, and the first the People team hears about it is an awkward Day 2 conversation.
Every pipeline below is a shape you wire on the canvas using the crew and tools further down. Not a feature we ship for you, a pattern you configure.
On inbound application, parse the resume, pull GitHub, arXiv, and LinkedIn context, then score against the structured rubric loaded as Static context so the scoring stays consistent across the loop and across weeks.
Search co-authorship networks, GitHub OSS, and LinkedIn for passive matches per open role. HITL gate blocks every linkedin_send_connection_request and gmail_send until the recruiter approves target list, copy, and cadence.
Benchmark base, bonus, and token grant against current crypto and quant comp bands pulled from rag_retrieve, flag any gap below market P40, and draft the offer letter in Docs. Send stays HITL until the hiring manager signs off.
Assemble the offer letter, IP assignment, equity grant, and GDPR notices for the candidate's jurisdiction from the policy handbook in the knowledge store. HRComplianceLegal flags missing clauses with replay on every revision.
Draft the 30-day performance-concern memo or PIP from manager input, prior 1:1 notes pulled via rag_retrieve, and the ER decision framework in Static context. Notes stay HITL and never auto-send to the employee.
Score every employee against the 14-signal attrition model, name the primary and secondary driver, and recommend the intervention. Cross-run memory tracks intervention milestones so save plans do not reset each Monday.
Real personas from the hr_team crew. Each ships with a tuned system prompt and a default tool allowlist. Swap models per persona on the canvas.
Runs the recruiting funnel for quant and tech roles, scores candidates against a structured rubric, and flags any pipeline where screen, interview, or offer conversion falls below threshold.
Predicts attrition at a 90-day horizon from 14 signals, surfaces the primary driver per employee, and sizes the hiring pipeline against expected departures.
Benchmarks every role against crypto and quant market data, sizes equity refresh, and lists every employee below market P40 as a flight risk.
Documents performance concerns within 30 days, runs PIPs to standard, and triages conduct and harassment cases on a 24-hour clock.
Tracks the multi-jurisdiction compliance calendar, screens contractor classification, and audits GDPR HR data processing with documented legal basis.
Maps the 12-month skill gap, builds role-specific curricula, and measures L&D effectiveness by 90-day performance lift rather than completion rate.
Tracks eNPS, regression-adjusted pay equity, and inclusion indicators, and flags the three culture risks that bite hardest during scale.
Sets people strategy, watches the four weekly metrics, and names key-person and team-health risks before they turn into business risk.
Synthesizes every workstream into a board-ready people report with a People Health Score, putting critical risk in sentence one.
Every tool below is a real shared tool from the Melaya bundle. Allowlist per agent; HITL-gate the writes; revoke any of them in one click.
Search LinkedIn for the right candidates at target firms, resolve profiles into dossiers, and stage outreach. Every linkedin_send_message and linkedin_send_connection_request is HITL-gated by default.
Read inbound applications and replies, prepare candidate and employee outreach, and stage every send for approval. gmail_send is HITL by default for this crew.
Pull public context from GitHub, arXiv, conference pages, and prior employer sites into the candidate dossier. Scoped at the agent level so TalentAcquisitionLead reaches only what you allow.
Build the per-workflow knowledge store from interview rubrics, leveling guides, comp bands, the policy handbook, and won-candidate emails. Powers all three knowledge layers across the crew.
Read the offer-letter and onboarding templates from Drive, append candidate-specific clauses, and write the weekly attrition and pipeline panels into Sheets. Drive and Sheets writes are HITL-gated by default.
Find common free slots across an interview loop, prep the structured-interview agenda, and stage the invites. gcal_create_event is HITL-gated by default.
General-purpose research, structured fetch, and the explicit pause_for_human checkpoint used at every HITL gate. The fallback when nothing more specific applies.
Every pipeline ships with three layers of knowledge access. Mix and match per agent on the canvas. No shared vector space with another tenant, no surprise reads, no opaque retrieval.
includeContextPer-pipeline documents appended to specific agents' input on every run. The ICP brief, playbook, pricing sheet, or won-deal email corpus. Whatever needs to be there before the agent thinks. You pick which personas get which docs.
rag_retrieveA scoped tool granted per-agent. When the agent decides it needs more depth, it queries the workflow's vector store on demand. Same knowledge base as Static context, accessed only when the model asks for it.
pipeline_memoryPipeline-level state that carries from one run to the next. Yesterday's research is in scope for today's follow-up. The crew remembers what it already prospected, what got approved, what was sent. The audit log is the second-order knowledge base.
Sandbox tier is free with no card. Join the waitlist and we will email you the moment a slot opens.
← Back to every use case