Bengaluru's H2LooP raised $2 million in seed funding from Speciale Invest and 3one4 Capital. They're building AI for the notoriously difficult, unseen embedded software that runs critical hardware in cars and drones. Traditional AI models struggle here, making hardware code a manual, months-long bottleneck for complex systems.
How We Got Here
Founded in May 2025, H2LooP aims to tackle the deep, often overlooked problem of low-level firmware and driver development. CEO Sairanjan Mishra (ex-Philips, Cisco) and CTO Pulkit Agrawal (ex-Google) built the product from their shared experience in complex systems.
The Numbers
- H2LooP was founded in May 2025 by Sairanjan Mishra (ex-Philips, Bosch) and Pulkit Agrawal (ex-Google).
- Their solution combines small language models trained on embedded code with a proprietary knowledge graph mapping hardware specifications and safety standards.
- The AI platform verifies generated code and can deploy on-premises or air-gapped, crucial for defense and semiconductor clients.
- H2LooP is already deployed with automotive and semiconductor teams in India and Europe, accelerating design updates for AUTOSAR compliance.
What Happens Next
🇮🇳 Why This Matters for India
For deeptech founders and embedded engineers in Bengaluru and Pune, H2LooP shows how specialized AI can unlock long-standing bottlenecks in core hardware and chip design.
The Take
The $2 million reveals a critical market gap for deeply specialized AI that reasons over hardware constraints and compliance. H2LooP's proprietary knowledge graph offers the key differentiator here, a capability general LLMs currently cannot replicate.
Source:
YourStory ↗