Hopper vs ml-intern

Hopper, from YC S24 startup Hypercubic, is the first agentic development environment purpose-built for mainframe systems. It lets AI agents navigate TN3270 terminals autonomously, write and submit JCL jobs, monitor JES output, debug failed jobs by analyzing spool data, query VSAM datasets, compile and run COBOL code, and manage CICS transactions—all via natural language prompts. Tasks that traditionally took mainframe specialists hours of manual TN3270 navigation can now be expressed as a single instruction. The technical challenge here is real: mainframes don't have nice REST APIs or modern dev tooling. They run on green-screen terminal protocols from the 1970s, and the humans who know how to operate them are retiring faster than they can be replaced. Hopper essentially wraps the entire mainframe interaction surface in an agent-friendly interface, translating intent into the arcane sequences of keystrokes and JCL that mainframes actually require. The product is free for individual developers (all core features, macOS/Windows/Linux) with Enterprise pricing for SSO, on-prem deployment, and SOC 2 reports. Hypercubic's team includes alumni from Cognition, Apple, and Windsurf. Given that mainframes still process an estimated $3 trillion in daily commerce and the COBOL developer shortage is acute, Hopper is targeting a genuinely underserved market with unusual urgency.

ml-intern is Hugging Face's own open-source autonomous ML engineering agent. Given a task description, it reads relevant papers, writes training code, executes it in a sandboxed environment, evaluates the results, iterates, and ultimately uploads a trained model to the Hugging Face Hub — with no human in the loop beyond the initial prompt. Under the hood, the agent runs an agentic loop of up to 300 iterations, using Claude as its reasoning backbone alongside smolagents. It has integrated access to HF documentation search, paper retrieval, GitHub code search, and sandboxed Python execution. When the context window fills (at 170k tokens), it auto-compacts rather than failing, and full sessions are uploaded to HF for inspection and reproducibility. What's notable here isn't just the capability — it's the source. Hugging Face is essentially shipping a proof-of-concept that the job of "write the ML training script, run it, fix it until it works, upload the result" can now be delegated to an agent. With 688 stars and active development as of this week, ml-intern is HF eating its own dog food on autonomous AI engineering. The "doom loop detector" that flags repetitive tool-use patterns is a candid acknowledgment of how agentic loops fail in practice.