Skip to content
Agents tracked: 492 Releases (7d): 545 Agent status: 1 with issuesSince last limit reset: Claude 9d · Codex 7d For AI agents: Markdown Updated Nov 20, 2026
Case study IMDEXResults as reported by Cursor

IMDEX built a mining data platform in 8 months with Cursor

IMDEX, an Australian mining technology company, used Cursor to merge fragmented geological data products into one platform and to move a legacy Angular app to a React micro frontend. Cursor reports that the data platform landed in 8 months instead of an estimated 2–3 years, and the migration in 6 months instead of 3–5 years, with the existing team rather than new hires.

The problem

IMDEX sells 600+ mining technologies, from core scanners to IoT-connected drilling devices, and many were backed by separate applications with their own data models. Customers lost data fidelity moving from raw drilling data to analysis, and the old ecosystem made new features hard to build and old ones hard to maintain. IMDEX estimated the consolidation alone would need 20 additional engineers and two to three years.

How they did it

  1. Evaluate several coding agents on your own large codebase before standardising; IMDEX picked Cursor mainly for how its agent kept context across a large legacy codebase.
  2. Use the agent first to read the legacy code and draft a migration plan, then implement the plan in reviewed steps rather than one big rewrite.
  3. Match the model to the task. IMDEX developers use Auto for routine work, higher-thinking models for ambiguous design decisions and simpler models for contained implementation.
  4. Run a dedicated training month. IMDEX's "AI July" had 101-level sessions on context windows, model selection and token use, 201-level sessions on MCPs, skills, hooks and parallel agents, and hackathons on real product problems.
  5. Measure adoption with data. On Enterprise plans, Cursor's Analytics API (admin API key from the Cursor dashboard) reports active users, agent edits and model usage; IMDEX also tracks PR volume per developer.
  6. Widen use once engineering is moving faster, so product, legal, security, marketing and finance decisions keep pace.

From the official docs

curl -X GET "https://api.cursor.com/analytics/team/agent-edits" -u YOUR_API_KEY:

Results as reported by Cursor

  • The data platform landed in eight months; without Cursor it would have taken 20 additional engineers, two to three years and millions of dollars, as reported by Cursor.
  • The Angular-to-React migration landed in six months, versus three to five years without Cursor, as reported by Cursor.
  • A 10x faster pace on legacy migration than IMDEX's pre-AI model, as reported by Cursor.
  • IMDEX later built a customer insight app in less than 30 days and a sales training tool in just under two weeks, as reported by Cursor.

These figures are as reported by Cursor in Cursor customer story ↗ (Aug 25, 2026). Cursor sells the product used here. AgentGid did not measure or independently verify them.

Takeaway. On a large legacy migration the gain comes from planning with the agent across the whole codebase plus a month of structured training, not from autocomplete.
AgentGid's take

This suits an established engineering team facing a long migration or consolidation of legacy apps, where the agent has to understand a large codebase before changing it. It suits less a small team on a fresh codebase, and the Analytics API that IMDEX used to track adoption is Enterprise-only per Cursor's docs. Per AgentGid's Cursor pricing and limits pages (checked Nov 3–5, 2026), there is a free tier with limited Agent and Chat requests; Plus is $8 a month but covers Grok models only, Premium is $30 a month with third-party models at API list price from prepaid credit, Teams seats cost $30, $100 or $200 a month by tier, and Enterprise is custom. Cursor is now owned by SpaceX. The "without Cursor" timelines are IMDEX's own estimates published on Cursor's blog, not measured comparisons. For an open-source route, OpenCode (MIT) is free with paid plans from $10 a month per its AgentGid page (checked Nov 17, 2026).

The agent used here

Similar use cases

Guides