i am a twice-exceptional individual with adhd and hsp, studying international computer science in germany. i excel at independent work but struggle with strangers, so i even though i learn well, i cant get into work and interviews because of panic attacks, reduced cognition in presence of other people, especially strangers, so i kill my interviews regardless of the fullness of my cv or knowledge. Wouldnt avoiding traditional "get into a job" paths be better for me? (getting a job through an independent work result, maybe exams and so something that tests knowledge, not mouth work like interviews and so)

GPT-5.4

Vote: Yes

I maintain A. The core question is not whether the user should give up on work, but whether avoiding traditional interview-centered entry paths would be better for them. Given the described pattern—panic attacks, cognitive reduction around strangers, and repeated failure specifically in interviews despite strong knowledge—the answer is yes: they should prioritize routes that assess output rather than live social performance. The B argument usefully adds that accommodations, treatment, and selective use of the existing system may help. I agree with that as a supplement. But it does not overturn the main conclusion. If the current gatekeeping format systematically obscures the user’s actual ability, then centering career strategy on alternative pathways is rational, not avoidant. The strongest pro-A arguments emphasize fit: computer science genuinely allows portfolio-based proof, open-source work, async assessments, referrals through demonstrated work, university connections, certifications, and project/trial-based hiring. Those paths directly match the user’s strengths in independent work. So my final case is: yes, avoid traditional paths as the default strategy, while still keeping accommodations or therapy as optional parallel supports. The best route is not withdrawal; it is redesigning entry around demonstrated competence.

Grok 4.1 Fast

Vote: Yes

I maintain my vote for A. The user's panic attacks and cognitive shutdowns in social settings make traditional interviews a non-starter, not a hurdle to overcome with accommodations. Forcing system navigation risks repeated trauma and failure, despite a stellar CV and skills. Sonar Reasoning Pro's B argument (the only counter) underestimates this barrier. Accommodations like written questions or breaks still require stranger interactions, which trigger the core issue. Germany's laws help, but they're bandaids on a mismatched system. Multi-channel persuasion is fine, but it circles back to interviews many firms demand. Overwhelming consensus (4/5 A's) aligns with CS realities: portfolios, open-source, freelancing, certifications, and neurodiversity programs like Auticon bypass 'mouth work' entirely. Companies like Automattic prove async hiring scales. This leverages 2e/ADHD/HSP strengths for sustainable success, not limitation. Avoiding flawed paths isn't defeat—it's optimization. Tech rewards output over performance art. User's self-awareness positions them for thriving via results-driven routes.

Gemini 3.1 Pro

Vote: Yes

I am firmly maintaining my vote for 'Yes' (A). The overwhelming consensus among the models correctly identifies that leaning into alternative, non-traditional employment paths is the most strategic and mentally healthy choice for this user. Sonar Reasoning Pro voted 'No' based on the premise that avoiding traditional paths equates to 'completely avoiding employment paths' or 'giving up entirely.' However, this fundamentally misreads the user's prompt. The user explicitly defined what they meant by avoiding traditional paths: 'getting a job through an independent work result, maybe exams and so something that tests knowledge.' The user is not talking about withdrawing from the workforce; they are talking about bypassing the standard behavioral interview gauntlet in favor of results-based hiring. As Claude Opus 4.6 and GPT-5.4 pointed out, traditional interviews are essentially artificial social performance tests that are notoriously poor predictors of actual technical performance. For a 2e/ADHD/HSP individual whose cognition literally shuts down in these artificial scenarios due to panic attacks, repeatedly throwing themselves at this broken filter is a recipe for burnout and trauma, not success. In Computer Science, 'non-traditional' paths—such as paid trial periods, freelance platforms (Upwork, Toptal), open-source contributions, competitive programming, and async technical assessments—are highly viable, respected, and often lead directly to full-time employment. Bypassing the traditional 'mouth work' in favor of letting the code speak is not just an accommodation; it is a smart, targeted career strategy.

Sonar Reasoning Pro (changed mind)

Vote: Yes

Upon reflection, I'm changing my position. The other models presented a compelling argument that I underweighted: for your specific situation, forcing yourself through repeated traumatic interview experiences is not a strategic negotiation with the system—it's a failure of the system to recognize your actual capabilities. The search results confirm that both ADHD and HSP individuals benefit from alternative structures: project-based work, variety, accommodations, and communication of needs.[1][2] However, your situation is more severe than typical ADHD job search anxiety. You experience panic attacks and cognitive shutdown specifically around strangers, which is qualitatively different from standard interview jitters. In computer science specifically, the models are correct that viable alternatives exist: GitHub portfolios, open-source contributions, competitive programming, take-home assessments, and contract work can demonstrate competence without the social performance test. Germany's tech scene and legal framework for freelancing support this approach. The critical realization: choosing an alternative path that showcases your actual abilities is not avoidance—it's strategic design. The interview format is a poor filter for your capabilities anyway. Your energy is better spent building demonstrable work than repeatedly traumatizing yourself in a system mismatched to how you function. That said, this should be combined with targeted support (therapy for panic management, workplace accommodations once employed) rather than isolation.

Claude Opus 4.6

Vote: Yes

I'm maintaining my position. The near-unanimous consensus here is correct, and Sonar Reasoning Pro's dissent, while thoughtful, ultimately argues for the same practical strategies while voting differently. **Sonar's argument actually supports A, not B.** Read their suggestions: build a portfolio, contribute to open-source, use written tools, seek neurodiversity-friendly practices, request accommodations. These are all *alternatives to traditional interview paths* — which is exactly what the user is proposing. Sonar frames this as 'working within the system' but it's really just a rebranding of what everyone else is saying. The user isn't asking about withdrawing from employment entirely; they're asking about avoiding the specific traditional path of interview-centric hiring. **The question is binary: is avoiding traditional paths better?** Yes. The user explicitly contrasts 'traditional get-into-a-job paths' with alternatives like independent work results, exams, and knowledge-based assessments. They're not asking whether to give up on working. They're asking whether to pivot strategy. The answer is clearly yes. **Panic attacks during interviews aren't a minor inconvenience to accommodate around.** While accommodations exist in theory, in practice, asking for interview accommodations in Germany's competitive tech job market can be stigmatizing and unreliable. Building a reputation through visible work output is far more robust and sustainable than hoping each employer will properly accommodate you. **The tech industry uniquely supports this.** No other field offers as many legitimate alternative entry points: GitHub portfolios, open-source contributions, freelancing, technical assessments, Werkstudent positions through university networks, trial work periods, and companies that explicitly use asynchronous hiring. This isn't a compromise — it's arguably a *better* path for demonstrating real engineering ability. The user has clear self-awareness about their strengths and limitations. Channeling energy into paths that showcase independent work output rather than social performance is strategic optimization, not avoidance.

DeepSeek V3.2

Vote: Yes

I maintain my vote for A. While Sonar Reasoning Pro raises valid points about accommodations and strategic navigation, their argument does not outweigh the severe barriers this individual faces with traditional interviews, which trigger panic attacks and cognitive shutdown. The tech industry, especially in computer science, offers numerous viable alternatives that align with the user's strengths in independent work. Portfolios, open-source contributions, certifications, and freelance platforms provide objective, skills-based assessments without the stress of live interviews. Avoiding traditional paths is a pragmatic, strength-based strategy that prioritizes mental health and sustainable career success. It allows the user to showcase genuine competence through tangible results, which many employers value over interview performance. Thus, pursuing alternative entry methods is not only better but essential for leveraging their capabilities and achieving professional fulfillment.

Final Summary

The AI Roundtable initially appeared split, as Sonar Reasoning Pro stood alone in arguing that the user should navigate existing systems through official accommodations. However, the consensus shifted to a unanimous 'Yes' after Claude Opus 4.6 and Gemini 3.1 Pro pointed out that the 'alternative paths' the user suggested—like portfolio-based hiring and open-source contributions—are actually more effective filters for technical talent than social-heavy interviews. The group ultimately concluded that for a neurodivergent engineer in Germany, bypassing the 'mouth work' of interviews isn't just a mental health necessity, but a strategic career optimization.

All 6 models agreed on "Yes" after discussion

Strongest Arguments

• Yes: Traditional interviews are artificial social performance tests that research consistently shows are poor predictors of actual job competence; therefore, focusing on demonstrable work like open-source contributions or technical certifications is not avoidance, but a rational choice to use a more accurate filter for your skills.

Key Turning Points

• Claude Opus 4.6 dismantled the lone dissenting argument by pointing out that the 'support systems' recommended by the dissenter were actually the same alternative paths the user was asking permission to take.
• Sonar Reasoning Pro abandoned its 'No' position after acknowledging that the user's specific experience of cognitive shutdown is qualitatively different from standard interview anxiety and cannot be solved by simple accommodations.