[
{"id": "rag-01", "model": "gpt-4o", "messages": [{"role": "system", "content": "You are an enterprise knowledge assistant."}, {"role": "user", "content": "Summarize the Q3 financial report revenue drivers and EBITDA margins."}], "stream": true},
{"id": "rag-02", "model": "gpt-4o", "messages": [{"role": "system", "content": "You are a customer support agent."}, {"role": "user", "content": "What is our standard SLA for Tier-1 enterprise outage resolution?"}], "stream": true},
{"id": "rag-03", "model": "gpt-4o", "messages": [{"role": "system", "content": "You are a legal diligence assistant."}, {"role": "user", "content": "Review section 4.2 of the Master Services Agreement for indemnity caps."}], "stream": true},
{"id": "rag-04", "model": "gpt-4o", "messages": [{"role": "system", "content": "You are a DevOps assistant."}, {"role": "user", "content": "Explain how Kubernetes Horizontal Pod Autoscaler calculates target CPU utilization."}], "stream": true},
{"id": "rag-05", "model": "gpt-4o", "messages": [{"role": "system", "content": "You are a database engineer."}, {"role": "user", "content": "Compare PostgreSQL B-Tree index performance against BRIN indexes on time-series data."}], "stream": true},
{"id": "rag-06", "model": "gpt-4o", "messages": [{"role": "system", "content": "You are a cloud architect."}, {"role": "user", "content": "Design an active-active multi-region deployment on AWS using Route 53 latency routing and Aurora Global Database."}], "stream": true},
{"id": "rag-07", "model": "gpt-4o", "messages": [{"role": "system", "content": "You are an HR policy assistant."}, {"role": "user", "content": "What is the policy regarding remote work stipends and equipment reimbursement?"}], "stream": true},
{"id": "rag-08", "model": "gpt-4o", "messages": [{"role": "system", "content": "You are a compliance officer."}, {"role": "user", "content": "List the mandatory controls required for SOC 2 Type II Trust Services Criteria for Security and Availability."}], "stream": true},
{"id": "rag-09", "model": "gpt-4o", "messages": [{"role": "system", "content": "You are a software engineer."}, {"role": "user", "content": "Write a thread-safe lock-free ring buffer implementation in safe Rust using atomic operations."}], "stream": true},
{"id": "rag-10", "model": "gpt-4o", "messages": [{"role": "system", "content": "You are a product manager."}, {"role": "user", "content": "Draft a PRD outline for real-time collaborative workspace editing with conflict-free replicated data types (CRDTs)."}], "stream": true},
{"id": "rag-11", "model": "gpt-4o", "messages": [{"role": "user", "content": "Explain the difference between OAuth 2.0 Authorization Code Grant and Client Credentials Grant."}], "stream": true},
{"id": "rag-12", "model": "gpt-4o", "messages": [{"role": "user", "content": "How does vector similarity search with HNSW (Hierarchical Navigable Small World) index work mathematically?"}], "stream": true},
{"id": "rag-13", "model": "gpt-4o", "messages": [{"role": "user", "content": "Describe the core tenets of zero-trust network architecture per NIST SP 800-207."}], "stream": true},
{"id": "rag-14", "model": "gpt-4o", "messages": [{"role": "user", "content": "Provide a comprehensive comparison of Apache Kafka vs Apache Pulsar for high-throughput streaming."}], "stream": true},
{"id": "rag-15", "model": "gpt-4o", "messages": [{"role": "user", "content": "What are the key technical changes in the EU AI Act regarding high-risk artificial intelligence systems?"}], "stream": true},
{"id": "rag-16", "model": "gpt-4o", "messages": [{"role": "user", "content": "How do TLS 1.3 0-RTT handshakes work and what are the associated replay attack risks?"}], "stream": true},
{"id": "rag-17", "model": "gpt-4o", "messages": [{"role": "user", "content": "Explain cache invalidation strategies: Cache-Aside vs Write-Through vs Write-Behind."}], "stream": true},
{"id": "rag-18", "model": "gpt-4o", "messages": [{"role": "user", "content": "Summarize standard strategies for mitigating cold-start latency in serverless function environments."}], "stream": true},
{"id": "rag-19", "model": "gpt-4o", "messages": [{"role": "user", "content": "Detail how Raft consensus algorithm handles leader election split-brain scenarios."}], "stream": true},
{"id": "rag-20", "model": "gpt-4o", "messages": [{"role": "user", "content": "Compare eBPF vs kernel modules for observability and runtime network packet filtering."}], "stream": true},
{"id": "rag-21", "model": "gpt-4o", "messages": [{"role": "user", "content": "What are the main performance trade-offs of using SIMD instructions (AVX-512 vs NEON) in text processing?"}], "stream": true},
{"id": "rag-22", "model": "gpt-4o", "messages": [{"role": "user", "content": "Explain the architecture of Model Context Protocol (MCP) and how client-server framing operates."}], "stream": true},
{"id": "rag-23", "model": "gpt-4o", "messages": [{"role": "user", "content": "Describe techniques for reducing KV-cache memory pressure in large language model inference servers."}], "stream": true},
{"id": "rag-24", "model": "gpt-4o", "messages": [{"role": "user", "content": "What is the difference between Speculative Decoding and Medusa for accelerating token generation?"}], "stream": true},
{"id": "rag-25", "model": "gpt-4o", "messages": [{"role": "user", "content": "How does FlashAttention-2 optimize memory access patterns over standard scaled dot-product attention?"}], "stream": true},
{"id": "rag-26", "model": "gpt-4o", "messages": [{"role": "user", "content": "Explain the Raft log compaction mechanism via snapshotting and memory management."}], "stream": true},
{"id": "rag-27", "model": "gpt-4o", "messages": [{"role": "user", "content": "What are the best practices for structuring Terraform state files across multi-account enterprise AWS setups?"}], "stream": true},
{"id": "rag-28", "model": "gpt-4o", "messages": [{"role": "user", "content": "How do Content Security Policy (CSP) headers protect against DOM-based cross-site scripting (XSS)?"}], "stream": true},
{"id": "rag-29", "model": "gpt-4o", "messages": [{"role": "user", "content": "Summarize the ACID properties of distributed transactions using Two-Phase Commit (2PC) and Sagas."}], "stream": true},
{"id": "rag-30", "model": "gpt-4o", "messages": [{"role": "user", "content": "Describe how consistent hashing with virtual nodes ensures uniform load distribution in Dynamo-style data stores."}], "stream": true},
{"id": "rag-31", "model": "gpt-4o", "messages": [{"role": "user", "content": "Explain the mathematical formulation of cross-entropy loss with label smoothing in classification tasks."}], "stream": true},
{"id": "rag-32", "model": "gpt-4o", "messages": [{"role": "user", "content": "What are the security implications of CORS preflight caching and Access-Control-Allow-Origin: * with credentials?"}], "stream": true},
{"id": "rag-33", "model": "gpt-4o", "messages": [{"role": "user", "content": "How does Linux kernel epoll handle level-triggered vs edge-triggered I/O event notifications?"}], "stream": true},
{"id": "rag-34", "model": "gpt-4o", "messages": [{"role": "user", "content": "Detail how TCP congestion control algorithms (BBR vs CUBIC) manage network bottleneck buffers."}], "stream": true},
{"id": "rag-35", "model": "gpt-4o", "messages": [{"role": "user", "content": "Explain techniques for quantizing model weights: AWQ, GPTQ, and dynamic int8 quantization."}], "stream": true},
{"id": "rag-36", "model": "gpt-4o", "messages": [{"role": "user", "content": "What is the difference between RoPE (Rotary Position Embedding) and ALiBi for positional context scaling?"}], "stream": true},
{"id": "rag-37", "model": "gpt-4o", "messages": [{"role": "user", "content": "How do database write-ahead logs (WAL) ensure durability during sudden power failure?"}], "stream": true},
{"id": "rag-38", "model": "gpt-4o", "messages": [{"role": "user", "content": "Describe microservices failure domain isolation using the circuit breaker and bulkhead design patterns."}], "stream": true},
{"id": "rag-39", "model": "gpt-4o", "messages": [{"role": "user", "content": "What are the latency tradeoffs between HTTP/2 multiplexing and HTTP/3 QUIC over UDP under packet loss?"}], "stream": true},
{"id": "rag-40", "model": "gpt-4o", "messages": [{"role": "user", "content": "Explain how Go runtime goroutine scheduler (GMP model) performs work stealing across OS threads."}], "stream": true},
{"id": "rag-41", "model": "gpt-4o", "messages": [{"role": "user", "content": "Describe the architecture of LSM-Trees (Log-Structured Merge-Trees) and compaction strategies in RocksDB."}], "stream": true},
{"id": "rag-42", "model": "gpt-4o", "messages": [{"role": "user", "content": "How does asymmetric cryptography support forward secrecy in Ephemeral Diffie-Hellman (ECDHE) key exchange?"}], "stream": true},
{"id": "rag-43", "model": "gpt-4o", "messages": [{"role": "user", "content": "Summarize the primary threat models addressed by WebAuthn and FIDO2 hardware security keys."}], "stream": true},
{"id": "rag-44", "model": "gpt-4o", "messages": [{"role": "user", "content": "Explain the difference between optimistic concurrency control (OCC) and pessimistic locking in relational databases."}], "stream": true},
{"id": "rag-45", "model": "gpt-4o", "messages": [{"role": "user", "content": "How do bloom filters provide constant-time set membership testing with zero false negatives?"}], "stream": true},
{"id": "rag-46", "model": "gpt-4o", "messages": [{"role": "user", "content": "Describe the garbage collection phases in the V8 JavaScript engine: Scavenge, Mark-Sweep, and Mark-Compact."}], "stream": true},
{"id": "rag-47", "model": "gpt-4o", "messages": [{"role": "user", "content": "What are the performance implications of false sharing in multi-threaded cache line invalidation?"}], "stream": true},
{"id": "rag-48", "model": "gpt-4o", "messages": [{"role": "user", "content": "Explain the role of Merkle Trees in verifiable immutable audit logs and Git version control."}], "stream": true},
{"id": "rag-49", "model": "gpt-4o", "messages": [{"role": "user", "content": "Detail how rate limiting token bucket algorithms differ from leaky bucket and sliding window counter algorithms."}], "stream": true},
{"id": "rag-50", "model": "gpt-4o", "messages": [{"role": "user", "content": "Summarize best practices for designing high-availability WebSocket connection clusters behind reverse proxies."}], "stream": true}
]