SkillRep.

The problem

Skills management has become a board-level priority — enterprises need to identify current skill levels, plan for emerging capabilities, and bridge skill gaps as technology evolves — but the underlying skills data has not kept up.

The HR team's pain is real and recurring: identifying target skill levels for each role is hard ("what *should* a Senior Software Engineer at our company know?"), and keeping up with emerging technologies ("what new capabilities are coming that should be on our roadmap?") is harder still. Generic skill libraries are too broad and instantly stale; in-house taxonomies require ongoing analyst time that no one has.

The solution

SkillRep generates a company-specific Master Skills library on demand — not a static reference dataset, but a taxonomy tailored to the requesting enterprise's industry, operating model, and technology stack. HR / L&D submits company name, URL, and up to three job roles; the AI returns Skills + Proficiency levels structured for direct ingestion into the company's Learning Management System.

The roadmap extends the data model from taxonomy to ontology: capturing inter-skill relationships, prerequisites, and emerging-tech signals so that the system answers not just "what skills?" but "what is the path between today's capability and tomorrow's strategic requirement?"

How it works

V1 flow: HR/L&D user fills a form with Company name, Company URL, Job Role (up to 3) → form triggers an n8n workflow → AI runs Master Skills generation, then Skill-to-Role mapping with Proficiency assignment → structured output returns to the user.

Primary model: GPT-4o, selected because skill generation demands depth and nuance — multi-shot prompts with clear examples dramatically improved quality during prompt iteration. Quality > latency > cost: the task is backend-load, async-friendly, and consumed downstream by systems that don't need sub-second response. Edge handling covers nonsense inputs, company-specific abbreviations, URL/company-name mismatches, and uncovered industries.

Optional context fields enrich the output: reference companies (anchor the taxonomy to comparable orgs), operating model signals (matrix vs functional), technologies (specific tech stacks to weight in the skill set).

Who it's for

Internal users: HR and L&D teams at enterprises running upskilling, right-skilling, and workforce-planning initiatives. They own the workflow and need the skills layer to power downstream systems (LMS, performance, hiring).

Initial GTM: SkillRep is being used as the foundation to transform an in-house LMS from a traditional Learning system into a Skills platform — proving the data layer's value before expanding to external customers. Admin sets up the Skills platform, creates a Company-level Master Skills library, then the Skills + Proficiencies cascade to roles and individuals.

Why it matters

Skills management is the key frontier of HR & Learning SaaS — the enterprise that can't answer "what skills do we have, what skills do we need, and how do we close the gap" can't execute any meaningful workforce transformation. The skills management software market is at $455M–$755M in 2025 with double-digit CAGR, and the underlying $47B HR Tech market is shifting from systems-of-record to systems-of-intelligence.

SkillRep's strategic bet is API-first, role-contextualized skills data: the layer that every other HR-tech application will need but few are building deeply. The future ontology — interrelationships, prerequisites, emerging-tech mapping — is the moat that turns a one-time taxonomy generator into the skills intelligence substrate for an enterprise.

At a glance

Project

SkillRep

Built by

Devaiah Kattera

One-liner

API-first, company-contextual AI skills taxonomy generator — turns job roles into structured skill + proficiency definitions for HR and L&D teams.