Our takeaway: The challenge is no longer proving that Physical AI can work. It is proving that it can work reliably, repeatedly, and economically in the real world. We unpack all of this in our latest DeepTech Dispatch: https://t.co/pce56tS9f0 (12/12)
Humanoid robots are finally leaving the lab. BMW, Mercedes-Benz, Amazon, Tesla and others are actively running Physical AI deployments. Billions of dollars have flowed into the category. Yet one question remains: Why aren't these systems deployed at scale? 🧵 (1/12)
The startup opportunity follows directly from the deployment challenge. Some of the most interesting companies may emerge around:
- Behavioural data infrastructure
- Simulation & digital twins
- Deployment-focused robotics
- Edge AI & real-time control (11/12)
Voice AI breaks in the real world. Accents, interruptions, noise. Most teams are left manually sifting through thousands of calls to find out why.
@SuperBryn_AI is fixing that and we are a proud partner in their journey.
#83: #AIRadarDaily — @SuperBryn_AI
For any engineering team building voice AI, there is a quiet, deeply frustrating reality known as the "Production Gap". In a controlled demo, an AI voice agent works perfectly. But the moment it hits the messy reality of the real world — with thick accents, sudden interruptions, background noise, and anxious customers — it breaks. And when it breaks, teams are left flying blind. They are forced to manually listen to thousands of call recordings just to figure out what went wrong, turning brilliant engineers into full-time QA testers. It is a slow, exhausting cycle that keeps great ideas permanently stuck in the pilot phase.
SuperBryn is stepping in to fundamentally change this reality.
Founded by Nikkitha Shanker (@nikkithashanker) and Neethu Mariam Joy (@DrNeethuJoy), SuperBryn is an intelligent evaluation and observability platform that helps enterprises understand exactly why their voice AI agents fail and how to fix them.
Neethu and Nikkitha have built a deeply sophisticated platform rooted in over a decade of speech recognition research. SuperBryn provides full-stack observability across the entire voice pipeline. It traces failures through speech-to-text, reasoning, and text-to-speech, running automated stress tests on live production traffic instead of just safe, synthetic scenarios. If a bot fails to handle an interruption or misinterprets an accent, the AI automatically pinpoints the exact point of failure.
What makes SuperBryn so special is its profound empathy for the people building these systems. It completely removes the guesswork and the black box of voice AI. By offering real-time alerting and root-cause analysis on production calls, it gives developers their confidence back. Instead of reacting to frustrated customer complaints, teams can see exactly where an agent struggled and deploy intelligent fixes, turning unpredictable bots into reliable, enterprise-grade systems.
The market? Healthcare organizations, financial institutions, and global enterprises that want to deploy voice AI at scale, but simply cannot afford the risk of a single dropped call or compliance violation.
Nikkitha and Neethu are true change agents. They looked beyond the hype of building yet another chatbot wrapper and chose to solve the much harder, unglamorous problem: making AI actually work reliably in the real world. Watching our homegrown founders combine deep academic rigor with sheer entrepreneurial grit is a beautiful reminder of the larger, unfinished agenda we are all part of. They are giving momentum back to builders everywhere.
Let's celebrate the builders.
w/ @jaybharatingle & @dikshantjoshi
#VoiceAI #ProductNation
With over 450 million Indians spending upwards of 4.5 hours on their phones every single day, the "scroll" has become our default state. But why settle for mindless browsing? Eloelo transforms your daily habit into an interactive experience.
We’ve brought the magic of live entertainment and local culture to your fingertips, making every swipe count. Experience the shift from passive viewer to active player today!
In partnership with @Kalaari
#BeyondTheScroll #QualityScreenTime #InteractiveEntertainment #kalaaricapital #BharatKaDigitalMunch
We covered the Earth Observation opportunity in depth in our second edition of DeepTech Dispatch. The global EO opportunity, the data gap, the use-cases, and where the next wave of EO startups can build. Read the full edition:https://t.co/sWN2GnMLj3
GalaxEye, an Indian SpaceTech startup launched Mission Drishti. The world's first OptoSAR satellite, and India's largest privately built Earth observation satellite.
India's private space sector is not just launching satellites. It's building its own technology stack, owning the IP, and taking it global. Mission Drishti isn't India catching up to existing EO capability. It's India creating a new category. Congratulations to Team GalaxEye!
We covered the Earth Observation opportunity in depth in our second edition of DeepTech Dispatch. The global EO opportunity, the data gap, the use-cases, and where the next wave of EO startups can build. Read the full edition:https://t.co/sWN2GnMLj3
India's private space sector is not just launching satellites. It's building its own technology stack, owning the IP, and taking it global. Mission Drishti isn't India catching up to existing EO capability. It's India creating a new category.