Real OA, phone-screen and onsite questions candidates reported at OpenAI, Anthropic, xAI, Citadel, Two Sigma, Squarepoint, Meta, Amazon, Google and 128 more — every row tagged with the level and the round it was asked in. Synced hourly.
4,122 questions · 2,819 at Senior/Staff+ · 294 ML & research · 197 quant · 673 OA · 854 reported in the last 90 days · 395 free
▶ Practice on TrueInterview · 📝 Submit an OA · 🗣️ Interview reports · 📦 The data (CSV)
Rows: who you are. Columns: where you're interviewing. Each number opens that list, newest first. What exactly each one selects →
| Who you are ↓ · Where you're interviewing → | 🧠 AI labs | 📈 Quant firms | 🏛️ MAANG+ | 🌐 All companies |
|---|---|---|---|---|
| 🎯 Senior & Staff+ ≈ L5 / E5 and up |
308 | 124 | 830 | 2,819 |
| 🚀 Staff+ only ≈ L6 / E6 and up |
60 | 13 | 156 | 482 |
| 🔬 ML & Research Scientist MLE · Research scientist · AI engineer |
76 | — | 91 | 294 |
| 📊 Quant researcher & developer QR · QD · QT |
— | 194 | 8 | 197 |
| 🗂️ Every level incl. new grad & unclassified |
356 | 276 | 1,172 | 4,122 |
By round: 📝 OA (673) · 📞 Phone (2,109) · 🏢 Onsite (1,641) · 🏠 Take-home (10)
By format: Coding (2,427) · System design (700) · OOD / LLD (589) · SQL (297) · AI coding (109) · 🆓 Free to solve (395)
Share of each format among the questions reported at that level (a question counts once, at the most senior level it was reported at). Recounted every sync.
| Questions | Coding | System design | OOD / LLD | SQL | AI coding | |
|---|---|---|---|---|---|---|
| 🚀 Staff+ | 482 | 10% | 74% | 5% | 6% | 4% |
| 🎯 Senior | 2,337 | 58% | 13% | 17% | 10% | 3% |
| Mid | 1,004 | 75% | 4% | 15% | 3% | 2% |
| New grad & intern | 284 | 87% | 0% | 8% | 2% | 2% |
And by employer: how much of what candidates reported came from an OA round, and the format mix.
| Questions | From an OA | Coding | System design | OOD / LLD | |
|---|---|---|---|---|---|
| 🧠 AI labs | 356 | 9% | 40% | 25% | 24% |
| 📈 Quant firms | 276 | 36% | 74% | 6% | 16% |
| 🏛️ MAANG+ | 1,172 | 12% | 59% | 18% | 12% |
🧠 AI labs: OpenAI (144) · Anthropic (79) · xAI (38) · Perplexity (24) · Scale AI (21) · Harvey (20) · Microsoft AI (13) · Luma AI (6) · Moveworks (6) · Cursor (5) · Mercor (5) · C3 AI (4) · Cohere (3) · Mistral AI (2) · Together AI (2)
📈 Quant firms: Citadel (72) · Two Sigma (41) · Squarepoint (34) · Akuna Capital (31) · Point72 (29) · Optiver (23) · Hudson River Trading (18) · Jane Street (11) · Voleon (11) · Chicago Trading (5) · IMC (4) · DRW (3) · SIG (3)
🏛️ MAANG+: Meta (335) · Amazon (320) · Google (266) · Microsoft (186) · Apple (142) · Netflix (88)
🌐 + 97 more companies
ByteDance (255) · Uber (217) · Snowflake (128) · LinkedIn (116) · DoorDash (107) · Pinterest (101) · Oracle (95) · Stripe (95) · Salesforce (88) · Roblox (86) · Capital One (84) · Bloomberg (83) · Databricks (79) · Airbnb (78) · Coinbase (72) · Snapchat (69) · Waymo (62) · Atlassian (61) · Goldman Sachs (53) · Rippling (48) · eBay (45) · Tesla (45) · Robinhood (44) · Lyft (43) · Instacart (38) · NVIDIA (37) · Expedia (34) · Reddit (34) · IBM (33) · Ramp (33) · Verkada (32) · SoFi (30) · Applied Intuition (29) · Confluent (29) · Upstart (29) · Walmart Labs (29) · Intuit (28) · Datadog (27) · Shopify (26) · Cisco (25) · Figma (23) · PayPal (23) · JPMorgan (22) · WeRide (22) · Pinduoduo (21) · Tradedesk (21) · Yelp (20) · Rubrik (19) · Affirm (17) · Whatnot (17) · Dropbox (15) · Visa (15) · Square (14) · Vanta (14) · HubSpot (13) · Okta (12) · Nuro (11) · Palantir (11) · Boston Consulting Group (10) · CVS Health (10) · Airtable (8) · Airwallex (8) · Asana (8) · Discord (8) · Amplitude (7) · Glean (7) · Lead Bank (7) · Arista (6) · Axon (6) · Bobyard (6) · Circle (6) · GEICO (6) · MongoDB (6) · Tubi (6) · Wayfair (6) · ZipHQ (6) · Block (5) · Disney (5) · Faire (5) · Gusto (5) · OKX (5) · Rokt (5) · Sigmacomputing (5) · StackAdapt (5) · Thumbtack (5) · AMD (3) · Brex (3) · Onemain Financial (3) · Oscar Health (3) · StubHub (3) · Render (2) · Bitkernel (1) · Houzz (1) · Indeed (1) · Nclusion (1) · Siemens (1) · Unknown (1)
Every company, with Senior/Staff+ and ML counts →
Newest first · 🔥 reported in the last 14 days · 🆕 last 45 days · Level = the seniority the loop was reported at · a dash means not recorded. Every company, newest first →
All 4,122 questions, every company, newest first →
A Research Scientist loop at OpenAI is not a SWE loop at OpenAI. These write-ups take one role at one company: the rounds in order, what each one grades, and the questions candidates reported in it. Free to read.
| Company | Overview | SWE | ML engineer | Research scientist | AI engineer | Infra | Quant |
|---|---|---|---|---|---|---|---|
| 🧠 OpenAI | 📖 | 📖 | 📖 | 📖 | 📖 | 📖 | — |
| 🧠 Anthropic | 📖 | 📖 | 📖 | 📖 | 📖 | 📖 | — |
| 🧠 xAI | 📖 | 📖 | 📖 | — | — | 📖 | — |
| 🧠 Perplexity | 📖 | 📖 | 📖 | — | — | — | — |
| 📈 Two Sigma | 📖 | — | — | 📖 | — | — | 📖 |
| 📈 Akuna Capital | 📖 | — | — | 📖 | — | — | 📖 |
| 📈 Optiver | 📖 | — | — | 📖 | — | — | 📖 |
| 🏛️ Meta | 📖 | 📖 | 📖 | 📖 | — | 📖 | — |
| 🏛️ Amazon | 📖 | 📖 | 📖 | 📖 | — | — | — |
| 🏛️ Google | 📖 | 📖 | 📖 | — | — | — | — |
| 🏛️ Microsoft | 📖 | 📖 | 📖 | — | 📖 | 📖 | — |
| 🏛️ Apple | 📖 | 📖 | 📖 | — | — | 📖 | — |
| 🏛️ Netflix | 📖 | 📖 | 📖 | — | — | 📖 | — |
What candidates said happened in the room — newest first, AI labs, quant firms and MAANG+ ahead of the rest, at most two per company.
| 🎚️ Level on every row | Senior, Staff+ or Mid — the seniority the loop was reported at, so an L6 candidate is not reading new-grad OAs. |
| 🧭 Round on every row | OA, phone screen, onsite or take-home: practise for the round you actually have next. |
| 🔬 Research & quant tracks | ML engineer, research scientist and quant researcher/developer questions, cut out of the bank instead of buried in it. |
| 🔥 Fresh, automatically | Synced from the live catalog every hour; 🔥 and 🆕 expire on their own, so the top of every list is what is being asked now. |
| ▶ Runnable, not a screenshot | Every title opens the real problem in an editor; coding, OOD and SQL get a judged verdict. Free rows open without a paid plan. |
| 📦 Data, not just markdown | data/questions.csv and data/questions.jsonl: every row with its companies, round, level and role. |
Saw a question in an OA or a loop? Submit it — company, role, level, round and a short source-safe summary. No screenshots, no copied problem statements: see CONTRIBUTING.md. Rows are generated from the catalog, so please do not open a pull request against a table.
Built something on the data? A study planner, a Discord bot, a chart — open an issue with the link and we will list it here. data/questions.csv and data/questions.jsonl carry every row with its companies, round, level and role, refreshed hourly.
- Tech-Interview-Questions — every company's loop, interview-process guides and what companies are asking, counted.
- New-Grad-Opportunities · Internship-Opportunities — open roles, refreshed daily.
Questions are listed by title and metadata only; statements and solutions live on TrueInterview. Code in this repository is MIT-licensed.