A psychographic agent turned a student app into a scalable SaaS
with Sebastian & Johnny, Talent Scout
A psychographic agent turned a student app into a scalable SaaS
Show Notes
Sebastian (Seb) and his father Johnny are the co-founding team behind Talent Scout, a B2B SaaS platform that uses a six-dimensional psychographic model to help HR directors and hiring managers screen candidates beyond basic qualifications - evaluating analytical ability, creativity, ethical alignment, soft skills, and cultural fit at scale, with a full evidence trail and built-in compliance guardrails.
The path to Talent Scout ran through a different product entirely. Seb transferred to USC as a sophomore, felt profoundly lonely in a school of thousands, and decided to build something about it. The result was Juntos - Spanish for “together,” a nod to his Bolivian father - a student connection app powered by the same psychographic agent that now drives Talent Scout. Juntos reached 200+ users across five universities in its first three months, but the B2C consumer model proved too fragmented to scale into a venture-worthy business. The psychographic technology, however, was clearly valuable. Talent Scout is the B2B pivot that put it to work.
What Talent Scout Does: Psychographic Candidate Screening
The problem Talent Scout solves is familiar to anyone who has ever posted a job on LinkedIn or Indeed: within hours, there are hundreds of applicants, and no human team has the bandwidth to evaluate them meaningfully. Most AI screening tools filter by hard qualifiers - location, required credentials, years of experience - and leave the harder judgment calls about cultural fit and soft skills entirely to an overloaded recruiter.
Talent Scout goes further. Upload a job description, and the platform analyzes it to create an ideal candidate profile, identifies essential requirements and skills, and generates stress scenarios calibrated to the role - used to assess experience level and cultural alignment. That analysis produces a 10-question self-assessment for candidates, with three predetermined answer options per question. Crucially, there are no right or wrong answers - the self-assessment measures whether a candidate's responses accurately reflect who they are, not whether they can guess what the hiring manager wants to hear. The six-dimensional model weights traits differently by role: an analytical role gets more weight on analytical traits; a sales role gets more weight on assertiveness and interpersonal dynamics.
On the sourcing side, Talent Scout integrates with LinkedIn and Indeed to pull resumes automatically. At the end of the process, candidates are ranked and classified - and every ranking is accompanied by the specific evidence behind it, drawn from the self-assessment and the resume. Hiring managers get data to inform their judgment, not a black box score telling them who to hire.
The Colorado AI Hiring Law as Competitive Moat
Colorado passed legislation limiting the use of AI in employment screening - one of the first US states to do so. For most AI hiring tools, this is a compliance burden. For Talent Scout, it is a selling point. The platform was designed from the start with guardrails that the new legal environment now requires: no pure AI scores without evidence, no discriminatory filtering by gender or other protected characteristics, and a full audit trail that can be produced if a hiring process is ever challenged.
The distinction Johnny draws between Talent Scout and just dumping resumes into ChatGPT is important: a raw LLM will give you results shaped by whatever biases and hallucination tendencies are embedded in its training. Talent Scout's infrastructure - the six-dimensional model, the guardrails against protected class filtering, the evidence-based outputs, the audit log - is what separates a compliant, defensible screening tool from a liability. As more states follow Colorado's lead, that infrastructure becomes a genuine moat.
The Pivot: From B2C Student App to B2B HR Platform
Juntos launched publicly in January 2025 and grew to 200+ users across five universities within three months - proof that the product resonated and that the psychographic agent worked. The problem was the business model underneath it. B2C consumer apps in 2025 face a market in which users have more alternatives than ever and switching costs are near zero. Monetizing a student social app requires either massive scale (which requires significant capital to acquire) or a monetization mechanic that students will actually pay for (which is a difficult sell in a demographic notorious for price sensitivity).
The pivot question surfaced during investor conversations: where else can this psychographic agent create value? The answer was obvious once asked. The same technology that identifies shared personality traits and values to connect students is exactly what you need to evaluate whether a candidate's traits align with a job description and a company culture. B2B HR software has clear buyers, willingness to pay, measurable ROI (time-to-hire, quality-of-hire), and a well-understood sales motion. Talent Scout was the result of pointing the same engine at a market that would actually pay for it.
Early Traction: Network Before Product
Talent Scout's early clients came entirely through Johnny's professional network - people who had seen his work, trusted his judgment, and were willing to try something new because of that relationship. This is one of the most consistent findings across early-stage startup success stories, and Johnny makes it explicit: the right network matters more than having a great product for initial go-to-market. A great product with no distribution stays invisible. A good-enough product with warm distribution gets feedback, iteration, and early revenue.
The validation coming back from HR VPs and recruiters has been notably positive. Johnny describes this as the first time in decades of tech work that early users do not say “but” - they consistently respond with enthusiasm and a request to move forward. That absence of the usual objection loop is one of the better early-stage signals a founder can get.
Seb's Founder Journey: Knowing When to Step Back
Seb graduated from USC in May 2025 with a degree in International Relations and Global Business. The months leading up to graduation were a study in the tension founders feel between the pressure to follow conventional career paths and the pull of something they have built. He received a job offer in New York City - his dream - and declined it. By the time the offer arrived, he had spent a month and a half in an obsessive job search spiral and had burned himself out before the opportunity even materialized.
His decision to step back from Juntos around graduation is worth noting for founders who are wrestling with similar moments. He did not declare it a failure or blame the team. He described it as “not the right time” - a feeling rather than a formal post-mortem. The team (developers Ben and Rappa) had been exceptional. The product had users. The market timing was simply off for the B2C model. Recognizing that and letting go without being consumed by it - rather than pushing past the point of diminishing returns - is one of the harder skills in early-stage company building.
Tools & Resources Mentioned
- Talent Scout - Sebastian and Johnny's B2B HR screening platform; apps.talentscoutassistant.com
- Juntos - the B2C student connection app that produced the psychographic agent technology underlying Talent Scout; launched January 2025, paused May 2025
- Psychographic Agent - the core AI model measuring personality traits, values, and attributes - originally built for student matching, repurposed for candidate-role alignment
- Colorado AI Hiring Legislation - state law limiting unrestricted AI in employment screening; Talent Scout was built to be compliant and uses this as a competitive differentiator
- LinkedIn / Indeed - sourcing integrations that allow Talent Scout to pull candidate resumes automatically into the screening workflow
- Reid Hoffman / LinkedIn - admired by Johnny for building a professional connection platform with a long-term vision of helping people rather than just monetizing attention
Frameworks
Six-Dimensional Psychographic Evaluation
Talent Scout scores candidates across six weighted dimensions - including analytical, creative, and ethical traits - with the weights calibrated to the specific role being filled. A technical role gets more weight on analytical traits; a client-facing role gets more weight on interpersonal and assertiveness dimensions. The weighting is derived from the job description analysis, not manually set by the recruiter.
Network Before Product for Initial Go-to-Market
Talent Scout's first clients came entirely through Johnny's professional network - not marketing, not inbound, not cold outreach. The lesson is that initial traction depends more on who you know and who trusts your judgment than on how good the product is. A warm network opens doors that a great product alone cannot. Build the network before you need it.
B2C to B2B Pivot on Core Technology
Juntos proved the psychographic agent worked. The B2C student app market was too fragmented to monetize at scale. The pivot to Talent Scout did not require rebuilding the technology - it required identifying a different buyer (HR director vs. college student) with a clear willingness to pay and a measurable problem. The underlying model was the same; the market was different.
Compliance as Moat
Rather than treating AI hiring regulations as a burden, design the product to exceed compliance requirements from the start. When regulations catch up - and in AI hiring, they are coming - a company that was already compliant has a competitive advantage over tools that were built without guardrails. Colorado's AI hiring law favors Talent Scout exactly because the platform was built with evidence-based outputs and protected class filtering from day one.
Not the Right Time as a Valid Startup Conclusion
Sebastian's decision to step back from Juntos illustrates an undervalued founder skill: recognizing when the conditions for a product to succeed are not present and choosing to preserve energy rather than push past diminishing returns. This is different from giving up on a good idea - it is honest assessment of market timing, business model fit, and personal readiness. The decision does not have to be permanent.
FAQ
What is Talent Scout and how does the screening process work?
Talent Scout is a B2B SaaS platform for HR directors and recruiters. Upload a job description, and the platform creates an ideal candidate profile, identifies essential requirements, and generates stress scenarios. This produces a 10-question self-assessment for candidates (three answer options per question, no right or wrong answers - the model measures accuracy of self-representation, not ability to guess the correct answer). Candidate resumes can be pulled in automatically from LinkedIn or Indeed. All candidates are ranked and classified with full evidence for each ranking.
What is a psychographic agent and how is it used in hiring?
A psychographic model measures personality traits, values, and behavioral attributes rather than credentials and experience. In hiring, the six-dimensional model analyzes a job description to identify which traits are most important for success in that specific role, then evaluates candidates against those weighted dimensions. The output is not a match/no-match decision - it is a ranked list with evidence, designed to give human hiring managers better data for their own judgment calls.
How does Talent Scout handle compliance with AI hiring laws?
The platform was built with compliance as a design constraint. Protected class characteristics (gender, race, age) cannot be used as filtering criteria - the system blocks those inputs dynamically. Every screening process generates an audit trail that can be produced for regulatory review. The Colorado AI hiring legislation, which requires transparency and prohibits fully automated AI hiring decisions, plays to Talent Scout's advantage because the platform was designed to empower human decision-making with evidence, not replace it.
Why did Juntos pivot to Talent Scout rather than continuing as a student app?
Juntos reached 200+ users across five universities in its first three months, validating the psychographic agent technology. The challenge was the B2C model: consumer apps require massive scale to monetize, and the student market is both price-sensitive and high-churn. During investor conversations, the question arose: where else can this psychographic technology create value? The answer - evaluating candidate-role fit for enterprise HR teams - pointed to a market with clear buyers, willingness to pay, and a measurable ROI. Talent Scout is the same technology aimed at a different, more monetizable buyer.
What does Talent Scout do that ChatGPT cannot?
A raw LLM will process resumes and generate rankings, but the outputs are shaped by whatever biases and hallucination tendencies exist in the underlying model, with no audit trail and no compliance guardrails. Talent Scout adds a structured six-dimensional evaluation model, protected class filtering, evidence-based scoring, and a full process log. In a regulatory environment where AI hiring decisions are increasingly subject to audit, that infrastructure is the difference between a defensible tool and a liability.
What did Seb learn from building Juntos as a student founder?
Several things stood out. First, surrounding yourself with the right people - open-minded, transparent, good listeners - matters more than almost any other factor. Second, you need to make sure your problem is not just yours - a product that solves a problem only you experience cannot scale. Third, knowing when to step back is as important as knowing when to push forward. Seb describes his decision to step away from Juntos not as failure but as honest timing assessment - the conditions for that specific product in that specific market were not right at that moment.