Explore Problems

The canonical pool of system design interview questions circulating in prep resources has not kept pace with what major tech companies are actually asking in 2024-2025. Candidates who prepare from top-50 lists encounter completely different questions in real interviews — domain-specific, time-sensitive problems like real-time fraud detection or collaborative sync. The mismatch wastes preparation time and creates false confidence.

AT&T customers face fraudulent billing disputes and unreturned phone credit investigations that consume 25 or more hours of their time with no resolution. The carrier's negligent handling of returns and billing errors crosses into harassment territory with no accountability mechanism. Consumers need better tools to document, escalate, and resolve telecom disputes without losing weeks of their lives.

Small loan providers charge undisclosed daily interest and include balloon payment terms not mentioned at origination, resulting in borrowers owing multiples of the principal amount. The information asymmetry is deliberate and systematic. Loan contract analysis tools and predatory lending pattern detection would help consumers identify these traps before signing.

Long-term cable subscribers face compounding annual price increases on individual fee line items — broadcast TV fees more than doubling over five years — with no proportional service improvement. Customers in areas with limited ISP competition have no leverage to resist these increases and are effectively captive. When competition finally arrives, customers switching away consistently find better pricing elsewhere.

Consumers discover debt collection entries on their credit reports from insurance companies for accounts they have no record of establishing. These collections typically arise from cancelled policies with residual balances that insurers send to collections without direct consumer notification. The lack of pre-collection communication leaves consumers with no opportunity to dispute or resolve the underlying balance before credit damage occurs.

A customer describes promotional pricing and guarantees that were promised at signup but never matched actual charges, with the bill rising significantly once promotional pricing lapsed after about three years. Customer service, including managers, could not clearly explain the resulting charges.

State Farm resists legitimate claim approvals, takes up to 8 months to complete authorized repairs, then cancels homeowner policies after a claim is filed — systematically punishing customers who use coverage they have paid for.

Angi and similar home services lead platforms charge service businesses $600+/month for lead subscriptions that produce no actionable work, then impose $1,000+ cancellation fees when businesses try to exit. The combination of unverifiable lead quality and financial lock-in traps contractors in subscriptions they cannot afford to keep or leave. This pattern is documented across Angi, HomeAdvisor, and Thumbtack.

T-Mobile applied a $13.34/month credit versus the $34.58/month documented in a written chat transcript, then charged a $70 restocking fee when the customer returned the device due to T-Mobile's own billing failure. Multiple escalations over two weeks produced no resolution. Customers with written documentation of promises still face the same stalling pattern.

State Farm pays third-party claimants from at-fault accidents multiples more than it covers for its own policyholders under PIP limits, leaving injured customers with six-figure medical bills despite carrying full coverage.

Small ecommerce businesses struggle to manage separate mobile apps, websites, and POS systems, creating operational fragmentation and high setup costs.

Infra engineers constantly switch between AWS Console, Terraform, terminal, and role management with no unified tool

Founders abandon SEO prematurely because the early feedback loop is slow and uneven before compounding begins

Indie developers repeatedly build products nobody needs because finding authentic unmet demand requires monitoring hundreds of community posts manually for genuine pain expressions

Deploying and managing AI models across multiple networked machines with varying VRAM/RAM requires manual configuration, lacks hardware-aware model selection, and has no built-in orchestration.

Zendesk slows during busy hours with multiple tickets, impacting SLA. Incomplete page translation forces manual Google Translate use.

GitHub Actions self-hosted runners fail to deregister when the runner process exits or crashes, causing orphaned instances to accumulate in Docker environments with restart policies until the runner limit is hit.

Developers building nutrition or food tracking applications find available food recognition APIs either prohibitively expensive for side projects, unreliable in accuracy, or so poorly documented they are unusable. This forces developers to abandon features or build their own pipelines from scratch. The gap leaves a large class of health and wellness apps unable to add viable food logging.

LoanCare extracts double the actual county-assessed property tax through escrow by applying a fraudulent administrative neighborhood overlay. The homeowner's county-assessed tax is $3,400 but the servicer charges $6,900 annually, pocketing the difference with no disclosure or justification.

Robotics engineers setting up GPU-based simulation environments (Isaac Sim, Gazebo, MuJoCo) face significant infrastructure overhead each time they start a new project or join a new team. The process of provisioning, configuring, and tearing down cloud GPU instances for headless simulation runs lacks any CI/CD equivalent, forcing teams to solve the same infra problems repeatedly. The pain is acute enough that teams starting fresh dread the ramp-up, even if they have solved it before.