> Markdown version of [/events/world-congress-2026-north-america/sessions/1677-you-can-t-rerun](https://www.wearedevelopers.com/events/world-congress-2026-north-america/sessions/1677-you-can-t-rerun). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# You Can’t Rerun Sunlight: Designing ML Data Architectures for Physical AI


- **Date:** Friday, Sep 25, 2026
- **Time:** 16:50–17:20 (30 min)
- **Room:** Stage 7
- **Event:** World Congress 2026 North America

## Recording

[Watch recording](https://www.wearedevelopers.com/videos/100389-you-can-t-rerun-sunlight-designing-ml-data-architectures-for-physical-ai)

## Description

Large language models are transforming how we build software, but physical AI systems expose a hard limit: you cannot recompute reality.

When robots, sensors, and production systems interact with the real world, failures are causal and time-based, not semantic. You cannot go back to record sunlight you missed, an observation you never captured, or robot telemetry lost to fragile connectivity. Backfills can repair a dataset while changing our understanding of what the system actually knew at the time.

In this talk, I will show how these constraints change how we design ML data platforms for robotics, agriculture, manufacturing, and other physical-world domains. We will look at how to capture irreversible data reliably at the site where it happens, preserve the difference between observed and reconstructed data, handle late and out-of-order events without rewriting history, and build a shared historical data backbone for analytics, training, and debugging.

I will close with where LLMs and agents fit into this workflow, and why they can reason about evidence but cannot recover evidence that was never captured.

## Speaker

### [An Phan](https://www.wearedevelopers.com/@an-phan)

Senior Data Infrastructure Engineer @ Hippo Harvest


## Related talks at this congress

- [From Simulation to Reality: Overcoming the Data Challenge in Physical AI](https://www.wearedevelopers.com/events/world-congress-2026-north-america/sessions/1787-from-simulation-to) — Mitesh Patel
- [Edge AI: Running Agentic Intelligence Where Internet Can't Reach](https://www.wearedevelopers.com/events/world-congress-2026-north-america/sessions/1805-edge-ai-running) — Nitin Eusebius
- [No Single Model to Rule Them All: Building Resilient AI Agents Across Open & Closed LLMs](https://www.wearedevelopers.com/events/world-congress-2026-north-america/sessions/1671-no-single-model-to) — Emmanuel Acheampong
- [Anatomy of an AI Request: Where Latency and Cost Are Really Born](https://www.wearedevelopers.com/events/world-congress-2026-north-america/sessions/1835-anatomy-of-an-ai) — Dan Fu