AI Threat Assessment · 26 May 2026

DSPWorks

IoT Hardware Engineering
STILL BREATHING
2.8/ 10

“DSPWorks has built something genuinely impressive: SIL-4 railway safety systems that must not fail, IoT modules hardened for the Indian operating environment, and the kind of deep RF expertise that survives first contact with a Mumbai car-shed roof. The problem is they've accidentally become a consulting firm cosplaying as a product company — brilliant at solving everyone else's hardware problems while their own product platforms remain 'coming soon' on a website that lists nine product families but ships engineering services.”

Business Model
3.5
Automation Risk
2.0
Moat Strength
2.5
Adaptability
4.0
Need Survival
1.5
AI Threat Level
BUSINESS MODEL REPLACEABILITY

Custom hardware engineering and safety-critical railway systems aren't getting automated by ChatGPT anytime soon — someone still needs to write MISRA-C code and survive RDSO certification. But their 'concept to product' consulting model is increasingly competing with AI-assisted hardware design tools and offshore engineering teams armed with Copilot.

3.5
WORKFORCE AUTOMATION RISK

Safety-critical firmware engineers who can navigate IEC 62279 compliance and railway certification are not getting replaced by Claude — the regulatory approval process alone creates a multi-year human bottleneck that AI cannot shortcut.

2.0
MOAT STRENGTH

RDSO certification for railway safety systems is a genuine regulatory moat — you cannot download that approval from GitHub. Twenty-five years of domain knowledge in Indian railway operating conditions, from Baramulla to Kanyakumari, creates real switching costs when your IoT module needs to survive coach vibration and work through metal roof shells.

2.5
AI ADAPTABILITY SIGNALS

Their careers page promises 'engineering culture and hiring process' details are 'coming soon' — the same phrase that appears across half their product portfolio. For a company building AI-resistant physical systems, they seem remarkably slow at building their own digital presence.

4.0
WILL THE NEED SURVIVE AI?

Physical IoT modules that must survive IP67, EMC compliance, and seven-year field deployments remain stubbornly analog problems. AI cannot debug why your LoRa antenna has a null at 1800 MHz or why your water level sensor drifts in chlorinated coach tanks.

1.5
Verdict

“DSPWorks survives because trains still need physical safety systems and someone has to debug why the RF front-end fails under real Indian operating conditions — problems that resist virtualization. They're the boring infrastructure company that everyone needs but no one disrupts; they just need to decide whether they're selling products or selling themselves.”

Scores are based on public information and AI analysis. This is an affectionate roast, not a financial assessment. The best companies use this as a mirror, not a verdict.

Roast another →