Delve accused of forking Sim.ai's open-source SimStudio and selling it as Pathways
Delve, a Y Combinator-backed compliance startup, faces new allegations that it forked an open-source tool and sold it as its own product without proper attribution or licensing. According to a TechCrunch report based on whistleblower claims, Delve pitched a no-code tool called Pathways to a prospect who later became the whistleblower known as DeepDelver.
Key allegations from the source
The whistleblower DeepDelver recognized that Delve's Pathways tool looked similar to Sim.ai's open-source agent-building product called SimStudio. When asked if Pathways was based on SimStudio, Delve representatives reportedly claimed they built it themselves. DeepDelver then presented evidence suggesting Pathways was actually a modified fork of SimStudio, changed just enough to be passed off as Delve's own work.
If true, this would violate the Apache software license, which requires proper attribution to the original developer. Sim.ai's founder and CEO Emir Karabeg confirmed to TechCrunch that Delve had no license agreement with Sim.ai. "We knew they planned to use Sim for something and later tried unsuccessfully to sell them an agreement," Karabeg said. "I didn't realize they were going to sell it out of the box as a stand-alone solution."
Additional context from the source
- Sim.ai was actually a Delve customer, with both companies being Y Combinator alumni
- Delve allegedly used these methods preceding its Series A funding round led by Insight Partners
- Insight Partners' 2025 blog post about its $32 million investment in Delve was temporarily unavailable from the VC firm's website
- Mentions of the Pathways tool on Delve's site appear to have been scrubbed
- Delve did not respond to TechCrunch's request for comment, and the media inquiries address on its website no longer works
This situation highlights the importance of proper open-source license compliance, particularly for companies selling compliance solutions. The Apache license requires attribution when using or modifying open-source software, and failure to comply can damage both technical and business reputations.
📖 Read the full source: HN LLM Tools
👀 See Also
llama.cpp Q8_0 quantization gets 3.1x speedup on Intel Arc GPUs with SYCL reorder fix
A fix to llama.cpp's SYCL backend brings Q8_0 quantization on Intel Arc GPUs from 21% to 66% of theoretical memory bandwidth, achieving 15.24 tokens/second versus 4.88 tokens/second previously on an Arc Pro B70 with Qwen3.5-27B.
UW Researchers Plan to Use Teacher-Worn Cameras for AI Training, Parents Opt-Out
University of Washington researchers planned to have preschool teachers wear first-person cameras to record children for AI model training, with an opt-out consent model.
Microsoft's BitNet Enables 100B Parameter LLM Inference on Single CPU
Microsoft's open-source BitNet project achieves 100B parameter LLM inference at 5-7 tokens/second on a single CPU, with the 2B parameter model using 0.4GB memory and 29ms latency while matching full-precision models on benchmarks.
Claude for Excel and PowerPoint Updates: Cross-Application Context and Skills Integration
Claude for Excel and PowerPoint now share conversation context across open files, with Skills available in both add-ins. The tools are accessible via Amazon Bedrock, Google Cloud's Vertex AI, and Microsoft Foundry for paid Mac and Windows users.