Stockholm-based startup Fika Jobs has raised $4 million to build what it calls a video-first hiring platform—one that leans on AI interview agents and short-form video profiles to reshape how candidates are discovered, screened, and evaluated. In a market where recruiting tools have often added friction (more forms, more steps, more “please upload your resume”), Fika’s pitch is refreshingly direct: make the first interaction feel closer to a conversation than a document exchange.
The company’s core idea is simple to describe and harder to execute well: replace or supplement traditional application materials with short videos, then use AI agents to conduct structured interviews that can be tailored to a role. The result is meant to feel like a hybrid of LinkedIn-style professional discovery and TikTok-style onboarding—fast, visual, and designed for attention spans that don’t naturally align with long job descriptions or multi-page resumes.
But the real story here isn’t just “video plus AI.” It’s about workflow design—how you turn candidate assessment into something that can scale without becoming generic, biased, or superficial. Fika Jobs is betting that the next wave of HR tech won’t be defined by dashboards or applicant tracking systems alone. Instead, it will be defined by interactive media and agentic automation: systems that can ask questions, follow up intelligently, and translate unstructured signals (like spoken answers on camera) into structured hiring outputs.
A platform built around two new primitives: video profiles and AI interviews
Fika’s approach centers on two components that work together.
First are short video profiles for candidates. Rather than asking people to submit a static resume and hope recruiters infer the rest, the platform encourages candidates to present themselves in a more human way—quickly, clearly, and in a format that can be consumed at speed. The company positions these videos as a more dynamic alternative to resumes, aiming to capture context that documents often miss: communication style, clarity of thought, motivation, and the ability to explain work in plain language.
Second are AI interview agents. These agents are designed to support parts of the hiring process, presumably including initial screening and early-stage evaluation. The promise is that employers can move faster through candidate pools while still maintaining a consistent interview structure. In theory, AI agents can standardize question sets, adapt follow-ups based on what a candidate says, and produce summaries that help recruiters compare candidates more effectively.
This combination matters because video alone doesn’t solve hiring. Video can easily become a popularity contest or a charisma test if it isn’t paired with a structured evaluation method. Likewise, AI interviews alone can become a “robot interrogation” if they aren’t grounded in role requirements and calibrated to avoid hallucinations or irrelevant questioning. Fika’s bet is that the pairing—video profiles for discovery and AI interviews for screening—creates a loop that is both scalable and more informative than either component by itself.
Why “video-first” is gaining traction in recruiting
Recruiting has always been a media problem disguised as an information problem. Employers need signal; candidates need opportunity. Resumes and cover letters are one way to transmit signal, but they’re also a bottleneck: they require interpretation, they compress nuance, and they often reward people who know how to write for ATS systems rather than people who can do the job.
Video changes the nature of the signal. It introduces tone, pacing, and the ability to communicate under light pressure—things that are difficult to infer from text. For roles where communication matters (customer-facing work, leadership, technical explanation, sales, product thinking), video can provide immediate value. Even for more technical roles, the ability to articulate tradeoffs and reasoning can be a strong predictor of day-to-day performance.
Still, video-first hiring has historically faced skepticism. Critics worry about bias (appearance, accent, gendered expectations), accessibility (people with disabilities or limited recording resources), and privacy (candidates may not want their likeness tied to employment decisions). The only way video-first platforms earn trust is by building guardrails: clear consent, options for accommodations, and evaluation methods that don’t reduce people to surface-level impressions.
Fika’s funding suggests investors believe those guardrails can be engineered—not just promised.
The unique challenge: making AI interviews useful without turning them into a gimmick
AI interview agents are where most recruiting startups either differentiate or fail. The difference between a compelling demo and a reliable hiring tool is whether the system can consistently produce outputs that recruiters trust.
There are several hard problems embedded in this:
1) Consistency across candidates
If the AI asks different questions to different candidates, comparisons become messy. If it asks the same questions to everyone, it risks missing context. A strong system needs a balance: a structured core interview with adaptive follow-ups that remain anchored to job requirements.
2) Calibration to role and seniority
A question that works for a junior engineer might be meaningless for a senior one. Similarly, a behavioral prompt for a customer success manager should differ from one for a data analyst. The platform’s value depends on how well it can map interview content to role competencies.
3) Handling ambiguity and incomplete answers
Real candidates don’t always answer perfectly. They ramble, they correct themselves, they misunderstand the prompt. An AI interviewer must be able to interpret intent, ask clarifying questions, and avoid penalizing candidates for conversational imperfections.
4) Avoiding “confident nonsense”
AI systems can generate plausible-sounding summaries that are wrong. In hiring, that’s not a minor bug—it can lead to unfair outcomes. The system needs mechanisms to ground its outputs in what the candidate actually said, and it should provide evidence or references where possible.
5) Bias and fairness
Even if the AI is “only interviewing,” it can still encode bias through question selection, scoring rubrics, or interpretation of responses. Video adds another layer of potential bias. Any credible platform must treat fairness as a product requirement, not a compliance afterthought.
Fika’s public framing emphasizes speed and dynamism, but the deeper question is whether the company is building evaluation logic that recruiters can audit. The most successful AI hiring tools won’t just automate interviews—they’ll make the results legible.
A broader shift: from applications to conversations
What makes Fika’s approach feel timely is that it aligns with a larger shift in how companies adopt generative AI. Many early deployments focused on productivity: summarizing documents, drafting emails, generating content. Recruiting is different. It’s inherently interactive and time-sensitive. Candidates are not static inputs; they are participants in a process.
By introducing AI interview agents, Fika is moving recruiting toward a conversational model. That model can reduce the time between “I applied” and “I’m being evaluated,” which is crucial because candidate drop-off is real. People lose interest when processes drag. They also lose confidence when they don’t understand what happens next.
A video-first profile can also improve transparency. If candidates can see what they’re expected to communicate—through prompts, examples, or structured video formats—they can prepare more effectively. That reduces the “mystery box” feeling that often surrounds early screening.
At the same time, conversation-based hiring raises new expectations. Candidates will increasingly expect responsiveness, clarity, and feedback loops. If AI interviews are used, candidates may want to know why they were asked certain questions and how their answers were interpreted. Platforms that ignore these expectations risk backlash, even if they technically “work.”
Where Fika could stand out: blending discovery and screening into one experience
Many hiring tools separate stages: discovery happens in one place, screening in another, scheduling in a third, and evaluation in yet another. That fragmentation creates delays and inconsistent experiences.
Fika’s concept—video profiles plus AI interviews—suggests a more integrated funnel. Candidates present themselves in a format that’s easier to evaluate quickly. Employers then use AI interviews to go deeper without immediately committing to human time. Recruiters can focus on higher-value interactions later in the process.
This integration could also improve data quality. When video profiles and interview transcripts are part of the same system, the platform can connect signals across stages. For example, if a candidate’s video profile indicates interest in a specific domain, the AI interview can tailor questions accordingly. If a candidate’s video shows communication strengths, the AI can probe for leadership behaviors or collaboration patterns. The key is that the system remains grounded in observable content rather than inventing narratives.
Investors backing this idea likely see the same thing: a chance to build a hiring workflow that feels modern to candidates and operationally efficient to employers.
The funding: what $4 million likely signals
A $4 million raise is not massive in absolute terms, but it’s meaningful for a pre-seed or early-stage company building a product that requires both engineering and iteration. Video-first platforms need infrastructure for media handling, storage, processing, and playback. AI interview agents require model integration, prompt and rubric design, evaluation pipelines, and safety controls.
That means the company’s near-term work is probably less about “proving AI can talk” and more about building a reliable end-to-end system: candidate onboarding, video capture guidance, interview orchestration, scoring and summarization, recruiter dashboards, and feedback loops.
It also implies the company is preparing for early employer pilots. In hiring, pilots are everything. The platform must demonstrate that it reduces time-to-screen, improves candidate experience, and produces outputs that recruiters can act on. Without that, AI hiring tools tend to stall after initial curiosity.
A unique take on “LinkedIn meets TikTok” (and why it can be more than branding)
The LinkedIn/TikTok comparison is catchy, but it can also be misleading if taken literally. LinkedIn is about professional identity and network discovery; TikTok is about entertainment and algorithmic feed consumption. Fika’s challenge is to borrow the best parts of both without importing the worst.
From LinkedIn, the platform can take structure: role alignment, competency framing, and professional context. From TikTok, it can take format: short, engaging, and optimized for quick comprehension. But hiring is neither purely social nor purely entertainment. It’s