Explore Problems
Showing 7,238 of 7,577 problems · matching your filters
Self-Improving AI Agents Are Inaccessible to Non-Technical Users
Running persistent self-improving AI agents requires Docker, VPS, and DevOps expertise, blocking non-technical users from the most capable AI systems.
Vocal Removal Tools Require Cloud Uploads Raising Privacy Concerns for Musicians
All major vocal remover and karaoke tools process audio via cloud servers, requiring users to upload potentially copyrighted or unreleased music to third-party infrastructure. Musicians, producers, and content creators handling sensitive audio files have no privacy-preserving offline alternative. This is a genuine gap for privacy-conscious users in the growing karaoke and music production market.
SCE billing disputes require CPUC escalation to get resolved
SCE billing errors persist for over a year through normal customer service channels, with resolution only happening within 48 hours of filing a formal CPUC complaint — indicating the standard support process is structurally non-functional.
Telecom cancellation channels all redirect to each other with no resolution
Customers attempting to cancel AT&T service find that physical stores refuse to process cancellations, online portals block self-service cancellation, and phone support transfers endlessly without resolution. The result is months of charges for a service the customer has actively tried to terminate through every available channel.
AI Agent Runtimes Mix Planning and Execution in One Layer
Node/TS agent code puts prompt assembly, model calls, tool routing, and persistence in one class. Makes testing, swapping, and moving workspaces painful.
Race condition in check-then-act spend-limit checks lets AI agents overspend
Many designs for gating an AI agent spending rely on separately checking a budget limit and then executing the spend, creating a time-of-check-to-time-of-use gap where concurrent or rapid actions can bypass the intended limit. This is a structural security and reliability flaw in how agentic AI systems that handle money or resource budgets are commonly architected.
Static rate limiters and circuit breakers require constant manual retuning
Rate limiters and circuit breakers work well only when tuned to current load and capacity, but that load, capacity, and latency drift over time, requiring ongoing manual reconfiguration to stay effective. The poster built an adaptive, self-tuning traffic governor (Levee) to remove this maintenance burden.
Carvana labels undisclosed clear-coat damage as normal wear and tear
A Carvana delivery driver pointed out cosmetic damage on a newly purchased vehicle that wasn't visible in the listing photos or 360-degree view, but Carvana denied the resulting claim by classifying deep scuffs through the clear coat as ordinary wear and tear. The customer also couldn't upload supporting high-resolution photos due to a broken upload feature on Carvana's site.
Phone carriers fail to apply promised trade-in credits for months
An AT&T customer was promised trade-in credits starting in February that, after repeated calls and multiple representative assurances, were still only partially applied by July, with the carrier continually pushing the resolution to the next billing cycle.
GEICO fails to add a driver to a policy, then bills for its own error
A GEICO customer was quoted a price to add her daughter's vehicle, but the company failed to also add the daughter as a driver, then raised the premium due to its own mistake. Multiple service representatives were unable to resolve the issue or provide a promised callback.
Insurance agents give conflicting coverage information that leads to denied claims
A policyholder was told by two different Allstate agents that their trailer and home had premium coverage tiers, only to have their claim denied nearly six weeks later after being routed through a shrinking local office with reduced staffing. The mismatch between what agents represent and what the policy actually covers directly led to a denied claim.
Indie SaaS founders struggle to find customers after launch
A founder built a complete SaaS product but got zero paying customers, concluding that distribution and customer acquisition, not product development, was the real bottleneck. This reflects a common structural gap for indie and early-stage builders who underinvest in go-to-market relative to building.
CSS and XPath selectors break easily when page structure changes
Developers building web scrapers and browser automation write CSS/XPath selectors that are brittle to page markup changes, causing scripts and tests to silently break. This has prompted community tooling (AI-assisted selector generation) aimed at producing more semantically resilient selectors.
HubSpot Enterprise Analytics Too Shallow, Advanced Features Priced Out of Reach
HubSpot Sales Hub analytics capabilities are insufficient for complex enterprise reporting needs while the advanced analytics features are gated behind pricing tiers that mid-market companies cannot afford. Growing companies fall into a pricing dead zone where SMB features are insufficient but enterprise pricing is unjustifiable. The gap forces teams to export data to external BI tools for basic reporting.
Aspiring Wellness Clinic Owners Lack Tools for Financial Feasibility and Setup
Non-practitioner entrepreneurs looking to open wellness clinics face a knowledge and tooling gap around startup cost estimation, practitioner compensation modeling, and regulatory requirements. Unlike franchise businesses, independent clinic setup has no structured guidance, leaving founders piecing together information from disparate sources. The combination of healthcare regulations, contractor versus employee decisions, and capital requirements creates a formidable barrier to entry without industry-specific resources.
Telecom falsely flags owned phone as stolen with no appeal path
AT&T suspended service on a legitimately purchased phone by placing it on a fraud blocklist, despite the customer possessing the device. After eight days of daily calls, multiple department transfers, and a filed fraud case, AT&T declined the appeal with no explanation. Customers have no independent escalation mechanism when telecom fraud systems produce false positives.
Stripe Documentation Hard to Navigate Between Related Concepts
Stripe's documentation, while comprehensive, is difficult to navigate when tracing how different concepts and APIs relate to each other. Developers implementing complex payment flows must repeatedly context-switch across disconnected doc pages. This navigation friction slows integration work and increases implementation errors.
Deferred interest charges ambush consumers after promotional periods end
Retail financing customers making minimum payments during promotional periods are hit with full retroactive interest charges when the period expires — a predatory pattern not clearly disclosed at point of sale. Consumers making good-faith payments have no warning before the deferred interest triggers. This structural deception affects millions of retail credit cardholders across major lenders.
Contractor Marketplace Confirms Bookings It Cannot Fulfill
Angi confirmed two separate contractor appointments in advance, then cancelled both within hours of the scheduled time. The pattern suggests bookings are confirmed without verifying actual contractor availability. Customers lose time and trust when confirmed commitments are repeatedly broken.
Reddit Penalizes Direct Link Posting, But Community Norms Are Not Clearly Communicated
Founders spend months posting product links on Reddit before discovering that link-dropping violates community expectations and the platform works on a community-first value exchange. Reddit provides no clear onboarding to the unwritten content norms that govern effective participation.