How other people’s “Round 3” maps onto Adarsh’s third live. Job 10454435 still none — no public IE names this Job ID for a live-round question. Unnamed stays unnamed. R1 (13 Aug 2026) and R2 (18 Aug 2026) already happened. OA is not a round. Java Live Code. Labels: IE-asked. Evidence: R3-R4-Question-Bible.md §18, _miner/PASTE-IES.md, and Question-Research-BIBLE.md. Other candidates reported these; your loop may differ.
OA-as-R1: if they numbered OA as Round 1, their “Round 3” is often this loop’s R2 analogue — already done. Flag every such remap. INTERN, US 3×60, FTC 5-round, SDE II HLD, experienced 4-round / ~5 YOE = NMF (not my format). Shiwangi bookstore OOD is already banked as a true third. Do not invent LeetCode IDs. Do not convert counts into 10454435 probabilities.
OA / coding test / workstyle ← not a round live R1 (13 Aug 2026) ← done live R2 (18 Aug 2026) ← done live R3 (next) ← may be DSA+LP, HM, GenAI+DSA, LLD, or BR possible R4 ← often Bar Raiser; sometimes extra tech if R3 was DSA
| Their heading | Maps to Adarsh |
|---|---|
| OA called Round 1, then “Round 2” coding | First live — done |
| Their “Round 3” / Tech Interview 2 after that | Second live — done, if the loop was two lives after OA |
| OA + two lives + later HM / BR / Fluency | Closest to your R3 / R4 |
| OA + coding + LLD + HM (three interviews) | Not UTA-default two-DSA. Rate Limiter lives here (count = 1) |
| Invite said Fluency, they paste a tree | Write the tree first (Distance K family). Essay after only if they still ask |
A third live is not always Bar Raiser, not always GenAI, not always more DSA. AUTA posts include HM after two DSA (LC 7563011, LC 8362604), BR as the third (LC 6806195, LC 7406809, Reddit 1qj304z), Fluency-as-DSA (LC 7623949), bookstore OOD (Shiwangi). Recruiter framing can be wrong: LC 7406809 told “3 LP only,” still got DSA. Full Parking-Lot OOD versus Trie/Heap DS-design: both appear in SDE I IEs; UTA Rank A is mostly two-DSA. Be ready for both. Do not claim one is “the AUTA R3.” HLD (Kafka, load balancers as a full system-design round) is not the SDE I UTA default.
Only loops where two live interviews already happened, then another live. OA omitted. Other candidates reported these; not a 10454435 list.
| IE | Their later slot | Maps to your third live? | What was in it |
|---|---|---|---|
| LC 7563011 AUTA BLR | HM 23 Oct after two DSA | Y — HM | Complex problem, POCs, multiple solutions; ~25 min; no DSA named |
| LC 8362604 AUTA APAC 2026 | HM 15 Apr after same-day two DSA; BR 30 May | Y — HM (BR = possible fourth) | Anagrams variation; Combine Garlands (ropes; overflow → long). BR: challenge, missing deadline, problem identified, simplifying |
| LC 6806195 AUTA | BR after R1 Mar + R2 May | Y — BR as third | Nice Subarrays (LC linked) + feedback / critical issues / beyond task |
| LC 7406809 AUTA | BR after Fri R1 + Mon R2 | Y — BR as third | Tree path A/B/C; extra space → LCA+DFS; couple LP unnamed. Recruiter said LP-only |
| Reddit 1qj304z AUTA Chennai | BR 2 Jan 2026 | Y — BR LP-only | LP only. Coding titles never posted on R1/R2 either |
| Naina UTA | R3 2 Jul after R1 12 Jun + R2 19 Jun | Y — LP + light SD unnamed | Two LP unnamed; light system design unnamed |
| Shiwangi UTA | R3 bookstore; R4 next day | Y — OOD as third | Bookstore word-count; classes. R4: work deep-dive + medium sliding window unnamed. Already banked |
| Reddit 1ueybmg AUTA | R3 HM/Gen-AI | Y | LC Hard sliding window unnamed; 2–3 general AI Qs unnamed. BR 40m behavioral after |
| Nisarg AUTA LinkedIn | BR 15:00 same day as two techs | Y — BR all-16 LP | STAR all 16; exact prompts unnamed |
| Divanshu AUTA LinkedIn | BR ~3 weeks later | Y — BR | Array/greedy medium 2-pass unnamed; in-depth LP unnamed |
| LC 7035289 AUTA | BR | Y — BR DSA | Unique Paths–like max points right/down only |
| IE.in 2024-grad AUTA | BR Feb 2026 after R2 4 Dec 2025 | Y — BR LP | ~37 min behavioral; critical feedback; complex discussion; deep dive; missed commitment; challenging implementation |
| LC 6653463 UTA | R3 | Y — LP + DSA | LP deep dive; negative feedback; longest substring all distinct characters |
| LC 6425074 | BR after consecutive-day lives | Y — BR LP-only | LP STAR only ~40 min; no DSA. Comp titles AUTA |
| LC 7623949 | Labeled GenAI Fluency | Y — Fluency is DSA | Nodes at distance K; O(n); no parent mapping. Then BR LP ~22 min |
| LC 7850431 | R3 GenAI+tech; R4 BR | Y | How use GenAI; should/should-not; dynamic k-th largest. BR: deep dive; didn’t know next; tough feedback |
| LC 7724048 | R3 HM; R4 BR | Y | Next Greater; LLM verify. BR: deep dive; missed commitment; convince someone. Aggressive Cows was their R2, not R3 |
| LC 8029194 | R3 HM | Y | Project ~30 min; Hard similar to Median of Two Sorted Arrays (similar, not stamped); YouTube prod issue 1 hour may use AI |
| LC 8014509 | R3 HM Gen-AI | Y | GenAI use/verify; LCA (LC linked); process scheduler LLD |
| Deepak LinkedIn | R3 | Y | Prompt engineering, token limits, DS for “highest KFC orders last 3 months” |
| Bhavya LinkedIn | Interview 3 HM | Y — HM | Spring login Controller/Facade/Service/Repository. Not Login Tracker. BR after: LP probing |
| GFG Apr 2026 selected | HM | Y | Significant technical challenge; limited information; conflict teammate/manager |
| GFG sde-1-17 | F2F3 | Y | LP extra for customers; client different POV; Validate sudoku (not Solver). R4: CN, transactions, Bankers, CAT→MEN (Word Ladder family, no LC id named) |
| GFG sde-1-34 | HM | Y | Two simple problems production-ready — unnamed titles |
| Arijit Char AUTA | Tech 2 unnamed SD | N — that was the second live | Do not file as your R3 |
| LC 6800853 AUTA | Their Round 3 grid/rewards | N — OA-as-R1; our R2 | Recruiter: cleared that slot, next is BR — that BR would be a third analogue if two lives already happened; the labeled “R3” DSA is not |
| Deepanshu Jindal AUTA | True R3 grid DP; BR RAG/feature-flags | snippet / LinkedIn 404 | Reliability: snippet. Grid DP as true third; BR discussion not a named DSA title |
| LC 6881058 | BR-only post | Possible fourth, earlier lives unverified | Maximum Sum Linked List from two sorted lists with common nodes. Do not stamp LC 962 |
I do not have AUTA-fresher evidence of a fifth live as a normal step. Absence is not a promise Amazon cannot add one. Receiving a third invite does not mean the loop is over. Several AUTA accounts schedule Bar Raiser after the third live. Mixed R3 also happens. Stay ready for Java.
| Source | They called Round 3 | OA-as-R1? | Your upcoming third? |
|---|---|---|---|
| GFG Off-Campus 2025 | Connect Sticks + Rotten Oranges | Y | N — our R2. Their BR after → our R3 analogue (role, deep-dive bug, conflict, negative feedback, learn outside job) |
| igreaper Medium | Rotten + Max Rectangle | Y | N — our R2. Their BR → our R3 analogue (multi-level why / impact / differently / disagreements / measure success) |
| Kamlesh LinkedIn | Stack Overflow LLD | Y | N — our R2. Their R4 HR/LP → possible R3 analogue |
| Prem LinkedIn | Their Round 3 DSA | Y | N — our R2. Their “R3 BR” Trapping Rainwater mod → our R3 analogue |
| Vaishali LinkedIn | Second onsite DSA as R3 | Y | N — our R2. Their GenAI then BR → our R3 then R4 |
| Older GFG “Round 3 (Technical Interview 2)” | Count set bits; BT to DLL; Vertical order; Sliding window max; Sudoku Solver; Trapping rain variation; Bottom view; … | usually Y or campus OA-as-R1 | N — our R2 analogue. See chapter 05 for pattern reuse, not as a third-live forecast |
| Vanshika | Interview 3 = Unix find | numbering odd | Not a clean two-DSA-then-third. Unix find mapped as second-coding analogue in the main bible — not your upcoming third |
Rate Limiter (IE.in 2025-grad) was their R2 of an OA+3 SD loop, not a third live, not UTA. Independent SDE I count remains 1. Their third was LP-only. LC 8362604 BR “rate limiting decisions” / geofencing were story follow-ups, not that LLD.
| Flag | Why | Example |
|---|---|---|
| INTERN | Typical intern loop is two lives, not a third after two FTE Live Codes. Fluency may be a min-heap, not Distance K | LC 7643209 intern Fluency = unnamed min-heap, not Distance K. IE.in oncampus intern Fluency = NGE variation + DP stocks as intern R1 |
| US 3×60 | Not AUTA APJ two-live then third | Levels.fyi oP6vow: merge two sorted int streams in their R3 |
| FTC 5-round | Not UTA two-live then third | LC 6629134 / 8368413 / 6567842; Roundz IE 172 LRU in R4/BR |
| SDE II / HLD | Rate Limiter HLD does not add SDE I count | LC 7311674; Hitesh; IE.in SDE-2 |
| Experienced 4-round / YOE | Different loop shape | DevBrainiac 126 ~2 YOE: their R3 Parking Lot is NMF; their R2 was labeled HM. Do not call Parking Lot “the AUTA third.” Parking Lot as SDE I R2 is GFG fresher — real, different slot |
| Different Job ID | Not 10454435 | Reddit 1s9z862 Job 3100855: R3 Burning Tree–like + Min Platform; BR Ownership / disagreement / failure / Customer Obsession |
| Canada AUTA | Geography | Jay Patel cart/checkout |
| Login Tracker | Unverified mentor prep | new_login / get_oldest_login — not a later-slot public ask. Closest public: LRU (Bhavya Interview 2, not R3) |
Format-similar intern IEs are difficulty-weaker. Do not treat intern Fluency as a prediction that your Fluency hour is a heap or is Distance K. LC 7623949 (FTE, Distance K, no parent map) and LC 7643209 (intern, unnamed min-heap) conflict if you collapse “Fluency.” Show the conflict. Write whichever they paste.
Candidate wording. Stamp an LC number only when the IE linked it. Unnamed stays unnamed. Rows whose IE heading was “Round 3” but remap to our R2 are omitted (trap table above).
| Wording | ID if named | Mapped | Source |
|---|---|---|---|
| Find All Anagrams variation | — do not stamp 438 | R3 HM AUTA | LC 8362604 |
| Minimum Cost to Combine Garlands (ropes; use long) | — | R3 HM AUTA | same |
| Count Number of Nice Subarrays | LC linked | BR as third AUTA | LC 6806195 |
| Tree path src→target print A/B/C; extra space → LCA+DFS | — | BR as third AUTA | LC 7406809 |
| Unique Paths–like max points right/down only | — | BR as third AUTA | LC 7035289 |
| Longest substring all distinct characters | — | R3 UTA | LC 6653463 |
| Distance K; O(n); no parent mapping | — | R3 labeled Fluency | LC 7623949 |
| Next Greater Element | — | R3 HM | LC 7724048 |
| Dynamic k-th largest (k changes) | — | R3 | LC 7850431 |
| Maximum Sum Linked List from two sorted LLs with common nodes | — do not stamp 962 | BR (earlier lives unverified) | LC 6881058 |
| DP on grids / path aggregation unnamed | — | true R3 AUTA (snippet) | Deepanshu |
| Longest Substring Without Repeating Characters | — | R3 HM | Ansuman |
| Hard similar to Median of Two Sorted Arrays | similar, not stamped | R3 HM | LC 8029194 |
| LCA | LC linked | R3 HM | LC 8014509 |
| Validate sudoku (not Solver) | — | R3 | GFG sde-1-17 |
| CAT → MEN one letter, dictionary intermediates | no LC id named | R4 | same |
| Nodes at distance K; LCA | — | R4 extra DSA | LC 6461439 |
| Distance between two nodes parent-only; then root parent = a leaf | — | BR as third | LC 6475219 |
| Most frequent 3-page sequence | — | BR as third | LC 6573582 |
| Reverse linked list in groups of k | — | BR as third | Uday |
| Modification of Trapping Rainwater | — | BR after OA-as-R1 two lives | Prem |
| Number of Islands; follow-up water→land largest island | unnamed follow-up | R3 (3-coding loop, not UTA two-DSA) | DEV 42i8 |
| LC Hard sliding window unnamed | — | R3 AUTA | Reddit 1ueybmg |
| Array/greedy medium 2-pass unnamed | — | BR AUTA | Divanshu |
| Medium sliding window unnamed | — | R4 UTA | Shiwangi |
| Binary Search solved using PQ | — | R3 | LC 7981646 |
| Bookstore — count of a particular word in a specific book | — | R3 OOD UTA | Shiwangi |
Not later-slot (do not remap as your third): Aggressive Cows = LC 7724048 R2. Connect Sticks / Rotten Oranges / LC 6800853 grid paths = our R2 under OA-as-R1. Celebrity = off-campus-10 R1 approach, not 2026 AUTA BR. Ansuman nearest-mall multi-source BFS = their R1. Intern Fluency min-heap = LC 7643209 intern first live.
Pattern families still worth warm Java: tree path / Distance K without parent map, heap (k-th largest, including dynamic k), grid DP (count paths, max reward), prefix/Nice Subarrays, Anagrams window, ropes/garlands with long, switch-at-common-nodes on two lists. Not a 10454435 probability matrix. Chapter 05 has the scan. Java cards for pasted later-slot titles: _answers/09-dsa-new-ies.md (max path between two leaves, remove k digits, Celebrity variant — paste-only until a URL). LC 6461439 is a different rejected loop than the Count-Inversions selected paste; do not merge.
Exact text or unnamed. Mentor STAR is practice only. Speak genuine cards from chapter 12 and _answers/05-star-br.md §3. Do not speak chapter 13 as yourself.
| Prompt | Mapped | Source |
|---|---|---|
| Complex problem, in-depth research, POCs, multiple solutions | HM AUTA | LC 7563011 |
| Challenge I solved; missing a deadline; problem I identified; simplifying a process | BR (fourth) | LC 8362604 |
| Feedback; critical issues; went beyond your task | BR as third AUTA | LC 6806195 |
| Couple LP unnamed “formality” | BR as third | LC 7406809 |
| BR LP only | BR as third | Reddit 1qj304z; LC 6425074 |
| STAR all 16 LPs; exact unnamed | BR same-day third | Nisarg |
| Current role; last deep-dive a bug; conflict coworker/manager; last negative feedback; learned something not required | BR after OA-as-R1 two lives | GFG 2025 |
| Describe a mistake you made. What did you learn? | BR | Rudraksh |
| Significant technical challenge; limited information; conflict teammate/manager | R3/HM | GFG Apr 2026 |
| Deep dive; missed a commitment; convince someone of your approach | BR fourth | LC 7724048 |
| Deep dive to find a problem; didn’t know what to do next; tough/critical feedback | BR fourth | LC 7850431 |
| Why that approach; what would you do differently; how did you arrive at that decision | BR | Rushikesh |
| How do you use GenAI; where should / should not; verify; efficiency | R3 / HM | LC 7850431; LC 7724048; LC 8014509; Vaishali |
| Ownership; Disagreement with senior; Failure; Customer Obsession | BR Job 3100855 NMF job-id | Reddit 1s9z862 |
| Two LP unnamed; light SD unnamed | R3 UTA | Naina |
Unnamed LP rows stay unnamed. Do not guess titles in the room. Default pair: Dive Deep (Ylogx −35% + three tiers) then Deliver Results (Horizon 17th/80+ unless intro already used the rover). Failure: Argus 73%. Deadline: do not invent a miss. Conflict: hybrid versus vector, or RLS versus app-only.
Shiwangi’s UTA loop is a true third after two lives. R1 2 Dec 2024, R2/R3 16 Dec, R4 17 Dec. Their R3 was bookstore OOD: count of a particular word in a specific book; classes, objects, methods. That is later-slot OOD, already in chapter 06 / the question bible. Their R4 was about 40 minutes work deep-dive plus a medium sliding window unnamed — do not invent an LC id. This is evidence that a third live can be OOD even when Rank A AUTA is mostly two-DSA. It is not evidence that your R3 will be a bookstore. Conflict to keep visible: “SDE I has no OOD” is false. “Parking Lot is the AUTA third live” is also false on Rank A two-DSA IEs. Parking Lot as their R3 is DevBrainiac NMF. Arijit unnamed SD was R2, not R3.
Prep both class design (locker / parking / bookstore-shaped) and HashMap+heap+trie DS-design. Rate limiter is not the default UTA R3.
This chapter is a remap. It is not a question list for Job 10454435. If a new paste arrives, put their heading in the left column, OA-as-R1 Y/N in the middle, your third live Y/N on the right, before you study the questions. Unnamed stays unnamed.