Members Lacking Provider Network Access
Reported by candidates from Oscar Health's online assessment. Pattern, common pitfall, and the honest play if you blank under the timer.
Oscar Health's April 2026 OA asks you to identify members without adequate provider network access. You're given providers with location and specialty, members with required specialties, and a max distance threshold. A member fails if even one required specialty has no provider within range. The trap is distance calculation and specialty parsing. StealthCoder will catch the edge cases when you're under pressure.
The problem
You are given a list of providers and a list of members. Each provider row is encoded as [providerId, specialty, x, y]. Each member row is encoded as [memberId, x, y, requiredSpecialties], where requiredSpecialties is a pipe-delimited string such as "Cardiology|Dermatology". A member lacks adequate provider network access if there exists at least one required specialty for which no provider of that specialty lies within maxDistance of the member. Use Euclidean distance. Return the ids of all members who lack adequate access in ascending order. Function Description Complete the function findMembersLackingAccess in the editor below. findMembersLackingAccess has the following parameters: Returns Member 101 has both required specialties within distance 5. Member 102 has no nearby cardiology provider. Member 1 is exactly distance 5 from the cardiology provider, so access is adequate. Member 2 requires dermatology, which is unavailable.
Reported by candidates. Source: FastPrep
Pattern and pitfall
This is a spatial lookup problem wrapped in data validation. For each member, parse the pipe-delimited specialties, then iterate through all providers to check if at least one of each specialty exists within maxDistance using Euclidean distance. The common miss: forgetting to check ALL required specialties, or mishandling the distance boundary condition (exactly at maxDistance counts as adequate). The problem gives you the logic flow in the examples. Build a helper function to calculate distance, then a nested loop to validate. Use a set to track which specialties are covered within range, then compare to required. If any specialty is missing, add the member ID to results. StealthCoder covers your math if the formula slips.
Memorize the pattern. If you can't, run StealthCoder. The proctor sees the IDE. They don't see what's behind it.
You can drill Members Lacking Provider Network Access cold, or you can hedge it. StealthCoder runs invisibly during screen share and surfaces a working solution in under 2 seconds. The proctor sees the IDE. They don't see what's behind it. Made by an engineer who treats the OA as theater. If yours is tonight, you don't have time to grind. You have time to hedge.
Get StealthCoderYou've seen the question.
Make sure you actually pass Oscar Health's OA.
Oscar Health reuses patterns across OAs. Made by an engineer who treats the OA as theater. If yours is tonight, you don't have time to grind. You have time to hedge. Works on HackerRank, CodeSignal, CoderPad, and Karat.
Members Lacking Provider Network Access FAQ
What does 'pipe-delimited' mean for specialties?+
It's a string like 'Cardiology|Dermatology'. Split on the pipe character. Each specialty must be matched exactly (case-sensitive, typically) against provider specialties. No fuzzy matching.
Does 'within maxDistance' include the boundary?+
Yes. The example states 'Member 1 is exactly distance 5 from the cardiology provider, so access is adequate.' So distance <= maxDistance is valid. Do not use strict less-than.
How do I calculate Euclidean distance between member and provider?+
sqrt((x1 - x2)^2 + (y1 - y2)^2). Compare the squared distance to maxDistance squared to avoid floating-point errors, or compare sqrt result directly. Either works if consistent.
What if a member has no required specialties or a provider has no specialty?+
The problem statement implies all rows are well-formed. Don't over-engineer for bad input. Treat empty specialty as a real value and match it literally. Focus on the happy path.
How do I optimize if there are thousands of providers?+
For a live OA, nested loops are fine if constraints are small. If they're huge, precompute a spatial index (KD-tree or grid). Start simple; optimize only if time limit fails. The problem likely expects brute force.