Explore Problems

Travelers find that individually priced deals on flights, hotels, or activities look attractive, but the real cost of a trip only becomes clear once every connected component and fee is added up, making genuine price comparison difficult.

A student reports frustration that study and exam-prep apps have poor interfaces and do not let them optimize their study plan around the metric that matters most: the exam date. This points to a gap between generic productivity tooling and students' actual scheduling priority.

People using AI to help format and rewrite documents currently can't see a clear diff of what changed, unlike code-editing tools that show line-by-line proposed edits before they're applied. Existing AI writing assistants tend to rewrite content wholesale, making it hard to selectively accept or reject specific changes within a document editor.

Investors managing projects from a distance need a reliable local project manager to act as their "boots on the ground," but finding and vetting someone trustworthy for this role is difficult. This recurring sourcing problem blocks remote real-estate investors from confidently starting or overseeing projects.

A Carvana customer received a vehicle with an already-cracked passenger windshield that was not disclosed at delivery, and Carvana refused repair despite internal records suggesting the crack had supposedly already been fixed before delivery. The customer cites multiple other reports from Carvana buyers describing similar undisclosed windshield damage, suggesting a recurring pre-delivery inspection or disclosure gap.

A customer paid $265 upfront through Wayfair for professional chandelier installation, but on install day the contractor demanded an additional $465 for scaffolding that wasn't disclosed or scoped beforehand, presenting it as non-negotiable with the installer already on-site. The customer ended up paying $730 total for a $265 quote with no chance to shop around or decline.

First-time rental property buyers are often unsure whether to hold title personally or through an LLC or other entity, and worry about overcomplicating a straightforward purchase. This structuring decision affects liability protection, financing, and taxes, and the lack of clear guidance leads to costly mistakes or analysis paralysis.

A Verizon customer reports that using the AI phone system to make a payment takes 30-40 minutes, while reaching a live agent to do the same task incurs a $10 fee. This reflects a broader pattern of carriers deploying slow AI phone support while charging extra for human assistance.

A Chase customer describes being passed between vaguely-named internal teams like escalation, research, and the Executive Office without resolution, eventually filing formal complaints with the OCC and CFPB. The account illustrates how large banks internal escalation structures can function as a maze that absorbs complaints without producing outcomes.

A customer whose kitchen cabinets were flooded and damaged during a paid appliance installation faced a third-party claims administrator that disputed the water damage, sent a fraud investigator, and offered a settlement far below the actual repair cost. The dispute illustrates how outsourcing installation-damage claims to third-party adjusters can leave customers without adequate recourse.

A T-Mobile customer was charged $192 for international calls that were actually made over WiFi via WhatsApp or under an active (and later cancelled) international plan, and support insisted the charges were valid despite the customer's evidence. Long hold times and a dismissive resolution process ultimately drove the customer to switch carriers.

A T-Mobile customer reports being repeatedly overcharged each month without consent, with promises from support to fix the charge next cycle only for the overcharge to recur. Frontline support is described as unable or unwilling to permanently correct the billing error.

Engineers using top-tier LLMs for mechanical design brainstorming find the models cannot reason accurately about 3D spatial relationships, clamping mechanisms, or manufacturability, producing plausible-sounding but physically implausible suggestions. This reveals a consistent gap between language-based reasoning and the spatial/physical reasoning needed for real hardware engineering work, regardless of which model is used.

Job seekers waste time applying to postings that are stale, unverified, or from employers who never respond, while employers get flooded with unqualified applicants. Existing job boards profit from application volume rather than match quality, leaving neither side accountable for wasted effort.

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.

Teams building AI agents on markdown-based context report recurring problems: context poisoning, internal contradictions, non-deterministic behavior, and difficulty navigating large context stores. This is a structural pain point in agent engineering as context volumes grow, prompting emerging structured-context-management approaches to replace ad hoc markdown dumps.

Teams that stand up a formal request-intake form still receive a steady trickle of work via direct messages because colleagues default to old habits. Those stray requests never enter the tracking system, so they are invisible to prioritization, capacity planning, and reporting. The gap persists even after the process change is announced and the form is live.

Users report that AI customer support agents sometimes mark a case as resolved and close it without verifying that the customer actual issue was fixed. This creates a risk of unresolved problems being silently dropped, undermining trust in AI-driven support automation.

Users of AI coding/chat assistants can lose weeks of accumulated context in a single mistake because most tools do not persist conversation history and working context locally. This drives builders to create ad hoc local memory systems to avoid repeating costly context rebuilding.

A developer running automated QA testing via Claude and an MCP server hits token/usage limits partway through a test run and has no way to persist task state externally so a fresh session or account can resume exactly where the previous one left off, forcing full restarts of long-running agent workflows.