ideas@asquared.uk
Levlr

Revolutionising talent discovery through AI technology

Levlr

Key outcomes

Revenue ahead of the curve

With active clients secured and income already generated, Levlr achieved results far earlier than most startups at the same stage.

Reduced hiring time and cost

By resurfacing overlooked candidates in existing talent pools, the platform delivers faster, more efficient recruitment.

Smarter, more inclusive hiring

Skilled candidates get a second chance, helping employers find talent they’d otherwise miss and strengthening teams in the process.

The challenge

Untapped talent in a broken system

In a recruitment landscape where 98% of candidates face rejection, hidden talent often goes undiscovered due to CV limitations and systemic biases. For the 12 million+ job seekers in the UK, this represents missed opportunities and a market that isn’t working as well as it could.

Traditional Applicant Tracking Systems (ATS) frequently filter out qualified candidates who don't fit the conventional mould. That includes recent graduates, career returners, those with hands-on skills but limited writing confidence, or people from non-traditional educational routes. Neurodivergent candidates, or those from cultures where self-promotion isn’t the norm, can also be disadvantaged thanks to rigid formats and keyword rules that ignore real potential.

Levlr approached us through our Impact Builder programme with an ambitious vision: to fundamentally disrupt this inefficient system by creating a platform that can identify overlooked talent, provide more equitable opportunities, and transform the recruitment process for both candidates and hiring managers.

The approach

From hypothesis to game-changing solution

Our collaboration with Levlr began with a critical hypothesis: “A significant percentage of rejected candidates possess skills not effectively communicated in their CVs." This perceived skills gap represented both a challenge and an opportunity to reconnect overlooked talent with employers seeking qualified candidates.

To validate the hypothesis, we realised the need to design and build a Proof of Concept (PoC) for the Levlr team. By conducting a technical feasibility study we’d be able to determine whether the concept was viable within the project’s constraints. 

As with any project, we began with a structured Discovery phase. In a collaborative workshop with Levlr, we explored the problem space and aligned on goals and messaging. This included creating proto-personas for candidates and recruiters to ensure user-centred design from the outset.

Since Levlr came through the Impact Builder, we also defined success from a business perspective, speaking with subject matter experts to better understand the market and gather unbiased views of the recruitment landscape.

Kickoff workshop with the Levlr team

Our engineers were involved from day one, working closely with our designers to blueprint the required technology. Given the platform’s complex data handling, the architecture evolved over several iterations to address emerging limitations. From here, we established an ambitious goal for our PoC: to resurface at least one interviewable candidate from the 98% rejection pool of candidates who would otherwise have been overlooked.

To do this, we developed a system that could:

  1. Analyse job descriptions and successful candidates to identify crucial traits and skills
  2. Establish benchmark scores based on previously successful applicants
  3. Generate match percentages for all candidates, including those previously rejected
  4. Create recruiter-friendly summaries highlighting relevant traits and skills

With a clear foundation in place, we mapped user flows for each persona, setting the stage for the final prototype and shaping the overall “Levlr experience”. Building on this foundation and leveraging their already strong branding, we focused our efforts on surfacing high-impact insights with minimal user friction, ensuring clear and immediate return on investment.

We then used real client data to test whether the system could identify qualified candidates from the rejection pool. And the results confirmed that AI could indeed surface overlooked talent who closely matched those who had progressed to interview.

The solution

An intelligent talent rediscovery platform

Building on the success of the PoC, we moved forward with developing an MVP web application that seamlessly integrates with an existing ATS. This integration was crucial as recruiters needed to maintain their existing workflows while gaining access to Levlr's enhanced candidate pool.

When recruiters log in, they immediately access recommended applicants for their open roles created in their ATS. The platform draws from a simulated talent pool including current and previously rejected applicants from similar roles, helping recruiters uncover candidates they might otherwise miss.

The science behind the match

At the heart of Levlr's platform lies its sophisticated and intuitive AI technology, which evaluates candidates on multiple dimensions:

  • Match score generation: Advanced natural language processing is used to compare candidate profiles against job descriptions through a carefully balanced evaluation system.
  • Intelligent benchmarking: The system establishes threshold values based on analysis of previously successful candidates, creating a reference point that helps identify promising applicants who might otherwise be overlooked.
  • Bias reduction:One of the platform’s key strengths is its use of anonymisation during evaluation, helping reduce bias and ensure candidates are assessed on skills and experience alone. The result is a fairer, more inclusive way to identify top talent.

To keep things clear and accessible for recruiters, Levlr presents matches in an intuitive Gold, Silver, and Bronze tier system. This approach streamlines decision-making without losing the nuance between candidates.

Seamless integration with existing systems

Understanding that adoption of digital products depends on the ease of use, we developed a public API that allows ATS to push job and candidate details directly into the Levlr database. 

The platform intelligently interprets applicant statuses from various ATS, allowing Levlr to remain system-agnostic while still accurately identifying previously successful candidates for benchmark scoring. This approach enables the platform to work across different recruitment systems without requiring extensive customisation.

The results

Early success

The impact of the Levlr platform has been remarkable. As a result, Levlr are on course to win investment and have already secured revenue from several active clients, far earlier than startups who are typically at the same stage in their growth cycle.

For hiring teams, Levlr offers more than just better candidates. By helping recruiters rediscover overlooked talent in existing pools, it cuts both time and cost. And, in a market where recruitment spend keeps rising, that efficiency is a major advantage. Levlr doesn’t just give rejected candidates a second chance, it helps prevent unnecessary rejections altogether.

A new era in recruitment

In a recruitment world full of inefficiencies and missed potential, Levlr marks a shift towards smarter, more inclusive hiring.

By combining AI with human-centred design, the platform opens up new opportunities for job seekers while delivering real value to recruiters. Our partnership with Levlr shows how thoughtful tech can transform the industry, not by replacing human judgement, but by enriching it with insights that would otherwise stay hidden.

Every resurfaced candidate represents a career given a second chance. And every organisation that makes a hire they might have missed gains not just a new team member, but a boost to performance, culture and future potential.

Request a quote

Hand drawn underline by a marker

Let's see what we can make together.

From startups to scaleups and enterprise, we're always happy to talk to impact-focussed and ambitious organisations. Use our quote creator to get a better idea of scope, timescales and next steps.

Request a quote
Update your cookie preferences