Start a research run
Give ENIAK a research topic. The dry-run loop fans out to a mock radar, extracts evidence cards through the model, and returns a citation-faithful chapter draft. Every claim points back to a source, and every call lands on the cost ledger.
Recent runs
7 runs- Succeeded
automated citation faithfulness in research books — Computational Methods for Faithfulness Detection (Leveraging Natural Language Inference models for claim-evidence verification; Retrieval-Augmented Generation strategies to ground citations in source documents; Evaluating semantic entailment versus surface-level keyword matching)
openai/qwen3.5-plus · May 27, 05:23 PM
- Succeeded
automated citation faithfulness in research books — Defining Faithfulness in Long-Form Scholarship (Conceptualizing citation faithfulness versus general factual accuracy; Quantifying hallucination risks in AI-assisted research book generation; Establishing baseline metrics for source-text alignment)
openai/qwen3.5-plus · May 27, 05:20 PM
- Succeeded
phase 5 e2e — multi-chapter book test — Designing Resilient End-to-End Flows (Architecture patterns for scalability and modularity; Risk mitigation strategies and failure mode analysis)
openai/qwen3.5-plus · May 27, 05:07 PM
- Succeeded
phase 5 e2e — multi-chapter book test — Foundations of Phase 5 E2E Systems (Defining scope, boundaries, and core objectives; Stakeholder alignment and requirement mapping)
openai/qwen3.5-plus · May 27, 05:02 PM
- Succeeded
env read at call time
openai/qwen3.5-plus · May 27, 04:30 PM
- Succeeded
post-deploy verification of citation invariant
openai/qwen3.5-plus · May 27, 03:55 PM
- Succeeded
supabase live e2e
openai/qwen3.5-plus · May 27, 03:44 PM