AI removed the speed bottleneck. We are the safety bottleneck it needs.
NirmIQ brings the discipline of safety-critical industries — automotive, aerospace, medical devices — to every team building software-defined and embedded products in the AI era.
Built by a 22-year automotive systems engineering veteran. Founded to make the toughest standards in engineering accessible to every team that ships safety-critical products.
Three shifts.
One intersection.
Speed is no longer the bottleneck
Copilots and AI agents now generate code faster than humans can review it. The constraint has shifted from “how fast can we ship?” to “how do we know it's safe?”
Products are software-defined
Cars, medical devices, industrial machines, electronics — today's products are mostly software running on embedded hardware. Legacy CAD-centric tools cannot govern this new world.
Every product is now safety-critical
A single bad software update at CrowdStrike caused $5.4 billion in losses. The standards developed for cars, planes, and medical devices apply to every team that ships software at scale.
The discipline of the few, brought to everyone.
For decades, the rigorous practices of safety-critical engineering — requirements traceability, FMEA, controlled change management, audit-ready documentation — were too expensive and too slow for most companies. Only Tier-1 automotive suppliers, aerospace OEMs, and medical device manufacturers could afford the people, the processes, and the legacy tools.
AI changed that. The same engineers who used to spend weeks writing requirements documents can now generate them from existing artifacts in minutes. The bottleneck is no longer effort. The bottleneck is discipline — making sure speed does not outrun safety.
NirmIQ brings safety-critical discipline to every engineering team — without the cost, the complexity, or the headcount that used to be required.
Madhusudhan Chellappa started his career as a Senior OBD Engineer at Cummins Engine Company in Columbus, Indiana, developing diesel exhaust aftertreatment and on-board diagnostics systems — including a novel DPF efficiency monitor still used on commercial trucks today.
He spent the next two decades working at the intersection of embedded software, control systems, and functional safety. At Clean Air Power in San Diego, he led product development for dual-fuel engine management systems integrated into Navistar, Caterpillar, and Volvo platforms. At Continental GmbH in Munich, he was the residential engineer managing system requirements for commercial vehicle programs across BRIC markets. At KPIT GmbH, he served as the lead consultant to GM Powertrain Europe, shaping decisions across $100M+ in powertrain development initiatives.
As Global Head of Embedded Systems at Geometric Ltd (acquired by HCL), he managed the global embedded systems practice for a $500M+ IT services company, defining product strategy for AUTOSAR, powertrain, and automotive testing solutions across teams in Germany and India. He secured multi-million-dollar contracts from top-10 global OEMs.
Most recently, as AVP at Genpact, he led a 250-person organization running a $60M automotive SaaS portfolio of 63 analytics and workflow products — including AI-assisted development practices.
Through every one of these roles, the same problem kept surfacing. The tools that govern safety-critical engineering were built for a different era. Requirements lived in one system. FMEA lived in spreadsheets. Code lived in Git. Auditors asked questions that nobody could answer without weeks of manual reconstruction. The smaller the company, the worse the gap — because the “real” tools cost six figures and took six months to deploy.
In 2016, he founded Gannet Engineering to fix that. NirmIQ is the result — built by someone who has actually shipped safety-critical products to global OEMs. Who has filed FMEAs for engines that go into trucks and cars. Who has sat through ASPICE and ISO 26262 assessments.
NirmIQ exists because the founder spent 22 years wishing it did.
A new category. Not a 1:1 replacement.
NirmIQ was not designed to replace IBM DOORS or Relyence one feature at a time. It was designed for what comes next — software-defined products built at AI speed, governed by the toughest engineering standards on the planet.
Built for software, not for CAD
Legacy PLM platforms (Dassault, Siemens, PTC) started from 3D mechanical models and bolted on software support later. We started from requirements, risk analysis, and code — and integrate to PLM through part numbers when needed. The natural fit for teams developing control software, electronics, calibration, and embedded firmware.
Requirements and risk are inseparable
Not a pure FMEA tool. Not a pure requirements tool. The most expensive engineering failures happen at the seam between "what we said we'd build" and "what could go wrong with it." NirmIQ treats requirements and risk as one continuous workflow.
Safety-critical standards, applied to everyone
ISO 26262, DO-178C, IEC 62304, AIAG-VDA — the toughest engineering standards on the planet. Designed because lives depended on the systems being built. Today, every company shipping software at scale has the same exposure that automotive Tier-1s have always had.
AI is the accelerator, not the threat
Other safety platforms treat AI as a competitor or a risk. We treat it as the engine that makes rigorous engineering finally accessible. AI proposes. The engineer decides. The system maintains the traceability. Speed and safety stop being a tradeoff.
The right thing.
Made the easy thing.
Make safety-critical engineering discipline accessible to every team that ships software-defined or embedded products — not just the ones that can afford a six-figure legacy deployment.
Use AI to compress the work, not to replace the engineer. The AI proposes, the engineer decides, and the platform maintains the traceability that auditors and regulators demand.
So the next CrowdStrike, the next Boeing 737 MAX, the next Therac-25 does not happen because someone could not fit the failure mode in a spreadsheet.
Built by people who shipped.
Not by outsiders with a hypothesis about engineering pain. By people who carried the work of safety-critical product delivery for global OEMs.
22+ years in automotive systems engineering. Senior OBD engineer at Cummins. Powertrain and ADAS at Continental, Clean Air Power, and KPIT. Global Head of Embedded Systems at Geometric/HCL in Munich. AVP at Genpact running a $60M automotive SaaS portfolio. M.S. Aerospace and EECS from University of Michigan. Founded Gannet Engineering in 2016.
Technology leader driving NirmIQ's platform architecture, AI capabilities, and cloud infrastructure. Focused on building scalable, enterprise-grade engineering tools.