The Mandate. The Intelligence. The Opportunity.
Hadaf's mandate is to serve three very different stakeholders. Hadaf's asset is 25 years of accumulated intelligence. The opportunity is to connect the two — deliberately, systematically, at scale.
Three stakeholders. Very different needs.
Clear needs are easy. Unclear needs are where intelligence matters.
25 years, already earned.
The intelligence exists. What's missing is the activation layer.
Transform 25 years of institutional knowledge and expertise into an AI Intelligence Hub — a reusable layer that powers agentic workflows across every stakeholder need, clear or unclear.
The AI Intelligence Hub.
Institutional assets become reusable layers. Layers power agents. Agents work with people. Every cycle gets smarter.
The foundation is what Hadaf owns. External signals are what keep it current.
Agents recommend. People decide. The system learns.
The Hub doesn't just serve. It evolves.
The Hub serves. And Hadaf drives its evolution.
Not a tool. Not a chatbot. An operating system that runs on Hadaf's own knowledge and expertise — extending it, activating it, and getting sharper with every cycle.
Traditional hiring ends. Agentic hiring compounds.
In a traditional hiring flow, if a candidate doesn't match a role, the process ends — and every insight collected during screening and interviewing is lost. Agentic hiring inverts this. Every candidate interaction enriches the institutional foundation. A missed match for Company A becomes a strong match for Company B — with a richer profile than before.
Two ways to run a hiring workflow.
The traditional flow ends. The agentic flow loops. That difference — repeated across millions of interactions — is how institutional intelligence compounds.
Seven modules. One connected workflow.
Every module below is a live capability in Lyzr's HR Agentic OS. Together they run the agentic flow shown above — with humans in the loop where judgment matters.
Pipelines pre-built for anticipated demand. When Nitaqat changes are forecast or sector demand surges, qualified Saudi talent is already identified and ready.
When a specific requirement arrives, the sourcing agent searches Jadarat, Tamheer graduates, Doroob completers — and applies 25 years of institutional patterns to surface non-obvious matches.
Ranks candidates on hard skills, culture fit, and predicted retention. Every ranking includes the reasoning — no black box. Officers see why each candidate scored where they did.
First-level qualification via voice interview in Arabic or English. Captures soft skills, motivation, communication style. Every response enriches the candidate's profile — permanently.
Officer reviews shortlists, interview summaries, and agent recommendations in one place. Approves, overrides, or requests refinement. Every decision logged and used to sharpen the agents.
Once the officer approves a shortlist, interview coordination is automated. Calendars synced, invitations sent, reminders managed. Officer time is protected for judgment work.
Placement activated. Subsidy pre-validated. GOSI registration triggered. Candidate onboarded to the employer's workflow. Outcome tracking begins — feeding the next cycle's learning.
The modules are ready. What Hadaf brings is the institutional knowledge that makes them uniquely yours.
Every build funds the next.
The AI Intelligence Hub is not built in one shot. It is built through a disciplined portfolio — measuring each candidate use case, then prioritizing across strategic value, human judgment, and shared capability. Every win funds the next.
Every workflow measured on quantitative criteria
Scoring produces a rank order. But rank alone doesn't reveal what to build first — or where to reuse what you've already built.
From scores to strategy
High value · High feasibility
High value · Lower feasibility
Lower value · High feasibility
Lower value · Lower feasibility
One capability, many workflows
The 2×2 does one more thing scoring can't — it surfaces workflows that share underlying capability requirements. Same data sources. Same integration patterns. Same agent architectures. Build the capability once. Reuse it across every workflow that needs it. That's how the portfolio stops reinventing the wheel.
Built once. Powers three workflows. Every future workflow needing finance data reuses it.
Without clustering, three workflows require three separate integrations — three teams, three timelines, three budgets. With clustering, one capability serves all three. And when the fourth workflow arrives, the same capability serves it too — at zero additional integration cost.
Small wins fund big builds. Shared capabilities compound. Momentum compounds into transformation.
Government-grade by design.
For a government partner, five things are non-negotiable: control of the stack, security at every layer, scale for national volumes, responsible AI at every decision, and full auditability. Everything else is a nice-to-have.
Non-negotiables for a government partner.
Any LLM. Any cloud. Any framework. Hadaf owns and controls the stack. No lock-in, ever.
SOC 2 Type II · ISO 27001 · PDPL-compliant · Data residency in KSA · Encrypted at rest and in transit · Zero data leaves the Hadaf boundary.
Proven at millions of transactions · Multi-tenancy · Horizontal scaling · High-availability SLA · Ready for national scale.
Bias monitoring · Fairness controls · Explainability · Human oversight at every high-stakes decision · Autonomy ladder from Assist to Autonomous.
Every agent decision logged · Immutable audit trails · Role-based access · Compliance reporting · Agent CI/CD with full traceability.
Sovereign full-stack agent platform
The Lyzr partnership — one platform, three enablers.
The Agentic Workbench powers the engine. The three enablers accelerate the journey on top of it.
Journey planning · Use case prioritization · Value realization · MBR/QBR frameworks
Applied engineers embedded with Hadaf · Accelerate first builds · Transfer capability to Hadaf's team
Enablement for Hadaf independence · Reusable modules · Faster future builds
The platform that powers everything above. Build, run, govern, and improve agents in one place. The three enablers accelerate delivery on top of this foundation. Without the Workbench, there is no engine to enable.
Lyzr is not a vendor Hadaf depends on. Lyzr is the enabler that makes Hadaf independent.
Proven where trust must be earned.
Government agencies. Global consulting firms. Fortune 500 enterprises. Lyzr has already been chosen where the stakes are highest and the scrutiny is deepest.
Trusted by the environments Hadaf operates in.
Strategic investors betting on Lyzr's agentic future.
Awards and analyst recognition.
By Mosaic Score. Top 6% of the cohort.
Named vendor in the emerging agentic AI category.
Agentic Technology category.
Recognized in the leading independent evaluation.
The credentials that matter.
Why partners choose Lyzr over building internally.
WTW went live faster on Lyzr than on internal build attempts. Where compliance is real, speed depends on the platform.
ShadowLM captures approved decisions, fine-tunes small models on them, and reduces inference cost over time — up to 90%. Internal builds get more expensive; Lyzr gets cheaper.
Audit trails, autonomy ladder, hallucination management, in-path guardrails. Building this internally is 5-6 months of engineering before writing a single workflow.
Hadaf brings 25 years of institutional knowledge and expertise. Lyzr brings the platform to activate it. Together — an AI Intelligence Hub that serves every stakeholder, every day, sharper every cycle.