Enginy

A named integration catalog is published with its own page and a logo wall carrying twelve more beyond those shown. The platform takes authenticated action rather than reading alone: bi-directional CRM sync writes verified contact and company data into HubSpot, Salesforce, Pipedrive and Zoho and keeps them updated in real time, outreach sends from connected mailboxes and social accounts, and reply alerts post into a Slack channel. Beneath the catalog sit thirty-plus enrichment providers feeding the data layer, which are inputs rather than action targets.

Sourceenginy.ai integrations page and CRM sync product pageread 2026-09-01

The platform carries the control flow and the branching is shown explicitly.

The sequence builder composes multi-step campaigns from named action nodes, send connection request, send message, like last social post, visit profile and send email, interleaved with wait steps of a set duration and with condition nodes, one of which is documented as testing whether a contact has accepted an invitation and routing accordingly.

Around it the platform orchestrates import, waterfall enrichment, list cleaning against ICP rules, outreach, reply handling and booking, and the developer documentation publishes a step-by-step reference for every Advanced Workflow import, enrichment and action step.

Sourceenginy.ai multichannel sequences and docs.enginy.ai Advanced Workflow referenceread 2026-09-01

Grounding runs on the prospect and on standing configuration rather than on a customer corpus.

Waterfall enrichment draws verified contact and company data from thirty-plus providers, AI deep research browses the web for profile and company insight per record, and list building is described as using real market signals together with the customer's company context, with an AI Playbook holding standing instructions the agents work from.

That context is authored configuration rather than a maintained retrieval structure: no document or file ingestion, no knowledge base and no retrieval interface over the customer's own material appears on the site or in the developer documentation.

Sourceenginy.ai AI powered research and find pagesread 2026-09-01

A review surface exists in one place. The smart inbox holds every conversation in one queue with a separate AI Drafts count, and the documented pattern is that AI prepares a reply which a rep can generate, edit and send, with intent tags prioritizing who to answer first and Slack alerts firing when a lead replies with interest.

The drafts are a convenience rather than a gate: the vendor also states the agent can keep conversations going autonomously and book meetings, and nothing lets a buyer mark actions as requiring sign-off, name who signs off, or define what happens while approval is pending.

Sourceenginy.ai smart inbox product pageread 2026-09-01

A security certifications section states the company is independently audited and certified to ISO 27001 and compliant with EU data protection rules, backed by a dedicated security policy page, a data processing agreement, a data opt-out page and a privacy policy in the site footer; the certification is asserted as held rather than in progress. No named access model states who inside a buyer's organization can do what, and no SSO, SAML or customer-facing audit surface is documented.

Sourceenginy.ai security certifications section and security policy pageread 2026-09-01

Tags conversations by intent with a stated reason and provides conversation and insight analytics. This is conversation and campaign analytics rather than agent execution tracing.

Sourceenginy.airead 2026-10-01

State persists across a long-running motion: contacts, companies and lists are held as standing objects in the workspace with enrichment written back, the smart inbox retains full conversation threads across email and social with intent tags and counts carried forward, an AI Playbook holds standing configuration the agents work from, and CRM sync keeps the durable record updated. No memory layer with a stated scope or lifetime is documented: nothing describes what an agent retains between campaigns, for how long, or a store it writes to and reads back distinct from the shared contact records.

Sourceenginy.ai smart inbox and CRM sync product pagesread 2026-09-01

Delivered only as a cloud SaaS platform; it is EU data protection compliant but offers no self host, on premises, or in VPC deployment option.

Sourcegenesy.airead 2026-10-01

Messaging and enrichment flows are configured per customer (ICP rules, sequences) rather than drawn from a library of prebuilt agents, templates, or packs.

Sourcegenesy.airead 2026-10-01

Work reaches the agents on signals rather than on a person asking: Enginy detects hiring changes, job transitions, tech stack updates and funding activity and starts outreach on them, and replies arriving in the smart inbox are categorized by intent and drafted for automatically.

Sourceenginy.airead 2026-10-01

The provider is disclosed rather than hidden: the message composition surface names the model in use as OpenAI GPT 5.2 alongside the personalization fields. That is a single named provider with no documented customer choice, no alternative models offered, no bring-your-own-key and no routing policy anywhere on the site or in the developer documentation.

Sourceenginy.ai multichannel sequences composition panelread 2026-09-01

A full developer documentation site is published at docs.enginy.ai covering four programmatic surfaces: a REST API for custom integrations with documented endpoints, authentication and guides for contacts, campaigns and webhooks; a hosted MCP server for connecting Claude or another AI assistant to an Enginy workspace; a complete reference for Advanced Workflow import, enrichment and action steps; and a specification for syncing an in-house CRM by implementing a small set of endpoints, with a reference implementation published in the vendor's GitHub organization. The docs subdomain is not linked from the marketing site's navigation or footer.

Sourcedocs.enginy.ai introduction and github.com/Genesy-AIread 2026-09-01

No user facing agent testing, debugging, or evaluation product is documented; messaging is tuned through the support team rather than an evaluation framework.

Sourcegenesy.airead 2026-10-01

The vendor documents its agent acting inside LinkedIn through named sequence steps: send connection request, send message, like a contact's last social post, and visit profile, with a condition node testing whether an invitation was accepted. Separately a Deep Research step is described as letting the AI agent browse the internet to find information, which is web retrieval wired into the product. No first-party page states how either the social actions or the browsing are executed.

Sourceenginy.ai multichannel sequences and AI research surfacesread 2026-09-01