> Markdown version of [/videos/1113-smart-connected-unexpected-the-wild-side-of-iot-and-ai](https://www.wearedevelopers.com/videos/1113-smart-connected-unexpected-the-wild-side-of-iot-and-ai). 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).

---

# Smart, Connected, Unexpected: The Wild Side of IoT and AI

How do you build AI for septic tanks and rat traps? Step into the wild side of IoT to master the rugged hardware behind bizarre industrial edge cases.

- **Speakers:** [Pawel Skiba](https://www.wearedevelopers.com/@pawel-skiba)
- **Event:** World Congress 2024
- **Published:** August 20, 2024
- **Duration:** 25:29
- **URL:** https://www.wearedevelopers.com/videos/1113-smart-connected-unexpected-the-wild-side-of-iot-and-ai

## Summary

Hardware development in the Internet of Things (IoT) presents physical and logistical hurdles quite distinct from software engineering, ranging from complex PCB component design and delayed manufacturing cycles to stringent electromagnetic compliance testing. Moving past off-the-shelf consumer gadgets, the most impactful IoT solutions often address unexpected, messy, and highly niche industrial problems. Tackling these bizarre challenges requires a resilient blend of rugged hardware engineering, strategic wireless connectivity, and applied artificial intelligence.

Deploying devices in real-world, hostile environments demands highly specific technical architectures. For instance, remote warehouse rodent monitoring leverages low-power, long-range LoRa networks to manage thousands of battery-operated traps in areas devoid of Wi-Fi or GSM. In industrial production, integrating edge computing with computer vision can automate the tedious process of identifying and counting insect species trapped on flypaper. Even subterranean applications, such as monitoring septic tanks, require specialized time-of-flight light sensors housed in acid-resistant enclosures to overcome the unpredictable, non-reflective surfaces of liquids.

Collecting training data for machine learning models in erratic physical spaces reveals the severe limitations of certain sensing technologies. While thermal cameras prove surprisingly GDPR-compliant and highly effective for tracking human movement in transit hubs or detecting pests inside building structures, they fail reliably at fever detection due to environmental fluctuations in skin temperature. Ultimately, successful hardware innovation relies on pragmatic problem-solving, whether that means manually soldering bypass adapters to prototype boards to avoid months of manufacturing delays or navigating the bizarre edge cases of human behavior during AI training operations.

**Keywords:** iot hardware development, pcb prototyping challenges, electromagnetic compatibility compliance, lora network deployment, edge computing computer vision, battery-powered remote sensors, time-of-flight level sensors, ruggedized iot enclosures, thermal camera people counting, gdpr-compliant motion tracking, fever detection limitations, machine learning training data, industrial pest control automation, wireless sensor algorithms, hardware manufacturing delays

## Chapters

1. **Complexities of hardware prototyping and compliance in IoT** (00:02) — Designing physical electronic boards involves prolonged manufacturing cycles, manual hardware debugging, and lengthy compliance testing.
1. **Automating remote pest tracking in large warehouse facilities** (04:19) — Contact sensors and low-power wireless networks enable remote monitoring of rodent traps across expansive physical locations.
1. **Streamlining insect identification in clothing manufacturing operations** (08:03) — Computer vision and edge computing reduce the manual labor required to count and identify clothing moths on glue traps.
1. **Overcoming sensory challenges in septic tank management** (10:46) — Creating reliable underground liquid measurement devices demands highly resilient hardware that withstands corrosive environments and connectivity dead zones.
1. **Repurposing thermal cameras for residential pest discovery** (14:34) — Nighttime thermal imaging successfully identifies roof-dwelling animals, allowing homeowners to resolve infestations without unnecessary extermination.
1. **Tracking human traffic with privacy-compliant thermal imaging** (17:19) — Using low-resolution thermal cameras over standard optical lenses ensures privacy regulation compliance while accurately tracking individuals in crowded areas.
1. **Navigating environmental variables in automated thermal fever detection** (20:53) — External temperatures and individual physical traits make surface-level thermal screenings highly unreliable for accurate medical diagnosis.
1. **Core takeaways from unexpected industrial hardware deployments** (24:23) — Real-world physical conditions regularly defy theoretical models, proving that automation opportunities exist even in unconventional legacy systems.

## Related Moments

- [Navigating hardware constraints and edge computing challenges](https://www.wearedevelopers.com/videos/1452-robots-2-0-when-artificial-intelligence-meets-steel) (from "Robots 2.0: When artificial intelligence meets steel")
- [Identifying authentic technology trends and news headlines](https://www.wearedevelopers.com/videos/1786-wearedevelopers-live-ai-freelancing-keeping-up-with-tech-and-more) (from "WeAreDevelopers LIVE – AI, Freelancing, Keeping Up with Tech and More")
- [Navigating AI integration limits in hardware and embedded systems](https://www.wearedevelopers.com/videos/100036-the-new-org-chart-when-ai-joins-the-workforce) (from "The New Org Chart: When AI Joins the Workforce")
- [Transitioning from web development to constrained hardware engineering](https://www.wearedevelopers.com/videos/557-iot-the-road-to-sustainability) (from "IoT: The road to sustainability")
- [High value in integrated systems and real world complexity](https://www.wearedevelopers.com/videos/100295-from-perception-to-autonomy-building-agentic-edge-ai-robots-with-ros-2) (from "From Perception to Autonomy: Building Agentic Edge AI Robots with ROS 2")
- [Questions on protocol performance, security, and industrial adoption](https://www.wearedevelopers.com/videos/771-from-bedside-beeps-to-interoperable-bytes-getting-your-medical-devices-ready-to-talk-the-ieee-11073-language) (from "From Bedside Beeps to Interoperable Bytes – Getting your medical devices ready to talk the IEEE 11073 language")

## Related Articles

- [Stephan Gillich - Bringing AI Everywhere](https://www.wearedevelopers.com/magazine/489-stephan-gillich-bringing-ai-everywhere)
- [Navigating the AI Shift](https://www.wearedevelopers.com/magazine/629-navigating-the-ai-shift)
- [Dev Digest 137 - AI'm not sure about this](https://www.wearedevelopers.com/magazine/485-dev-digest-137-ai-m-not-sure-about-this)
- [Dev Digest 138 - Are you secure about this?](https://www.wearedevelopers.com/magazine/486-dev-digest-138-are-you-secure-about-this)

## Related Jobs

- [Security Architect - AI](https://www.wearedevelopers.com/jobs/ext/1581899-security-architect-ai) at **ZEISS Group**
- [Senior Engineer, Infrastructure Platform](https://www.wearedevelopers.com/jobs/ext/328836-senior-engineer-infrastructure-platform) at **Intercom, Inc.**
- [AI Software Engineer (Germany)](https://www.wearedevelopers.com/jobs/48317-ai-software-engineer-germany) at **Sunhat**
- [Principal Software Engineer, Enterprise AI Platform](https://www.wearedevelopers.com/jobs/ext/1467292-principal-software-engineer-enterprise-ai-platform) at **GitHub**
- [Engineer, Offensive Security Organization](https://www.wearedevelopers.com/jobs/ext/1992296-engineer-offensive-security-organization) at **Twilio**
- [Principal Engineer - AI Search & Vector Infrastructure](https://www.wearedevelopers.com/jobs/ext/381484-principal-engineer-ai-search-vector-infrastructure) at **Redis**