# Flexcompute Engineering

> Essays, tutorials, and case studies on AI engineering, computational physics, photonics, and simulation from Flexcompute.

This site publishes essays, tutorials, and case studies from Flexcompute engineering.
Use the canonical article URLs for citation and the Markdown mirrors for lightweight machine ingestion.

## Feeds
- RSS: https://engineering.flexcompute.com/feed.xml
- JSON Feed: https://engineering.flexcompute.com/feed.json
- Sitemap: https://engineering.flexcompute.com/sitemap-index.xml

## Index Pages
- Home: https://engineering.flexcompute.com
- Articles: https://engineering.flexcompute.com/articles/
- Authors: https://engineering.flexcompute.com/authors/
- Tags: https://engineering.flexcompute.com/tags/
- Series: https://engineering.flexcompute.com/series/

## Articles
- Title: Learning Auto-Routing by Building: From Brute Force to an Auto-Design Agent
  Canonical URL: https://engineering.flexcompute.com/articles/electrical-routing-agents/
  Markdown URL: https://engineering.flexcompute.com/articles/electrical-routing-agents.md
  Published: 2026-04-20
  Updated: 2026-04-20
  Kind: Essay
  Authors: Prash Kharel
  Topics: AI Engineering, AI Agents, Photonics
  Summary: How I learned auto-routing for photonic chips — a failed brute-force attempt, AI as a learning partner, interactive HTML arenas, a PhotonForge router, and an agent that iterates 27 designs in under three minutes.

- Title: Predicting Peak Memory for an Electromagnetic Mode Solver
  Canonical URL: https://engineering.flexcompute.com/articles/mode-solver-memory-calibration/
  Markdown URL: https://engineering.flexcompute.com/articles/mode-solver-memory-calibration.md
  Published: 2026-04-15
  Updated: 2026-04-15
  Kind: Case Study
  Authors: Momchil Minkov
  Topics: Photonics, Tidy3D, Verification
  Summary: How we replaced a heuristic memory estimate with a calibrated model for Tidy3D mode solver workloads, eliminating under-predictions across the calibration set.

- Title: Can AI Agents Autonomously Design Components on Photonic Chips?
  Canonical URL: https://engineering.flexcompute.com/articles/autoresearch-photonic-design/
  Markdown URL: https://engineering.flexcompute.com/articles/autoresearch-photonic-design.md
  Published: 2026-04-13
  Updated: 2026-04-13
  Kind: Essay
  Authors: Tyler Hughes
  Topics: Photonics, Inverse Design, Optimization, AI Agents
  Summary: We gave AI agents a photonic simulator, a DRC engine, and four design challenges. They autonomously designed waveguide bends, crossings, splitters, and demultiplexers — some reaching near-perfect performance.

- Title: Designing a Photonic Chip Component with ~45 Lines of Python
  Canonical URL: https://engineering.flexcompute.com/articles/photonic-inverse-design-45-lines/
  Markdown URL: https://engineering.flexcompute.com/articles/photonic-inverse-design-45-lines.md
  Published: 2026-03-05
  Updated: 2026-03-05
  Kind: Tutorial
  Authors: Tyler Hughes
  Topics: Photonics, Inverse Design, Optimization, Tidy3D
  Summary: A compact introduction to photonic inverse design with Tidy3D, using a pre-built simulation and a ~45-line optimization loop.

- Title: "What Should We Work On Next?"
  Canonical URL: https://engineering.flexcompute.com/articles/what-should-we-work-on-next/
  Markdown URL: https://engineering.flexcompute.com/articles/what-should-we-work-on-next.md
  Published: 2026-02-26
  Updated: 2026-02-26
  Kind: Case Study
  Series: AI Engineering
  Authors: Yannick Augenstein, Frederik Schubert
  Topics: AI Engineering, Autodiff, Verification
  Summary: The story of building an 80,000-line autodiff library almost entirely through AI agents — and the verification infrastructure that made it possible.

- Title: The Agent Control Loop — Engineering for Tolerance
  Canonical URL: https://engineering.flexcompute.com/articles/agent-control-loop/
  Markdown URL: https://engineering.flexcompute.com/articles/agent-control-loop.md
  Published: 2026-01-19
  Updated: 2026-01-19
  Kind: Essay
  Series: AI Engineering
  Authors: Frederik Schubert, Yannick Augenstein
  Topics: AI Engineering, AI Agents, Verification
  Summary: Why agent reliability isn't magic model behavior — it's an environment where correctness is continuously verified. A framework for deciding when and how to delegate to AI agents.