Saurabh Datta Background / 2018-2022

Before consulting: R&D leadership at VW Group Innovation

Four years building and leading the Creative Technology R&D division at VW Group Innovation's Future Center Asia: patent-filed interaction hardware, shape-changing surface systems, user-trainable gesture recognition on edge ML, and the concept, creative direction, and full HMI architecture of the Gen.Urban concept-vehicle program.

IMG-ICA-01: one collage or single strong bench/buck photo from the ICA era
Year
2018-2022 / Beijing
Role
Built and led the Creative Technology division; concept authorship, creative direction, and engineering in one role
Stack
Interaction hardware / shape-changing surfaces, three generations plus pneumatic / edge-ML gesture systems / LED-matrix-behind-textile / Processing and openFrameworks HMI
Unlock
The practice the consulting now sells, run at institutional scale for four years

What the era produced

ADAS control stick PoC (2018)

Ergonomic mechanical design, electronics, and the simulation handshake I built personally; a patent was filed and I presented the concept to the Group CEO of the time, with encouragement to continue.

IMG-ICA-01 / ADAS control stick prototype and simulation rig

Gen.Urban V-0 and V-1 (2018-2020)

The deepest entry here, because it was concept, creative direction, AND build:

  • The concept itself was the contrarian bet: a reduced, minimal-information interior, the deliberate opposite of what was winning in China at the time (ever-bigger screens, 3D graphics everywhere). Interfaces recede when not needed and appear in context, so the cabin stays a calm space rather than a cockpit of displays. I carried creative direction of this overall concept alongside the HMI development lead role.
  • The "personalised interiors" universe, four components working as one: an interactive skylight with animated content; hidden displays behind fabric, where the LED-matrix-behind-textile research led to its own visual design direction and custom content tooling; ambient light at the foot rest; ambient sound. I personally built the enabling HW/SW experiments, wrote the guidelines the team created final content against, and built the management software binding the four together.
  • Shape-changing surfaces as part of the whole: the morphing side control panel (ground-up mechanics, electronics, and control software across three generations, plus the separate pneumatic soft-surface approach with its own toolchain; materials follow-up ran with Continental) delivered the same design philosophy physically, a control that exists only when context calls for it.
  • The hidden main display: a modular vertical-slider mechanism with its own configuration software, chosen (real display over AR) for content clarity and cost, designed so a remote workshop could mount and calibrate it without the builder present.
  • A machine-aesthetics UI direction: monochrome, terminal-style, an homage to old machines yet clutter-free and readable, developed with the freelancer ailadi (remotely, via shared boards), prototyped on real hardware in Processing and finalised in openFrameworks (C++).
  • Delivery under constraint: V-0 rescued a stalled concept in ~6 weeks of consolidation for the 2019 showcase; V-1 carried full HMI system responsibility for the 2020 showcase from Beijing under COVID, with a modular architecture that others assembled remotely from shipped documentation, illustrated technical riders, and per-component configuration tools (including the seat's CAN control module, specified from Beijing and built to instruction in Wolfsburg).

IMG-ICA-02 / Gen.Urban interior: skylight, hidden displays, morphing panel

IMG-ICA-05 / machine-aesthetics UI and delivery under constraint

Morphing surfaces, three generations (2019-2021)

See above for their role in the concept; as standalone R&D: full ground-up HW/SW including the pneumatic approach and its rapid-manufacturing toolchain, with Continental engaged on the elastic-optical material.

IMG-ICA-03 / morphing surface generations and pneumatic toolchain

User-trainable gesture systems on edge ML (2021)

Broad gestures via wearable TinyML, fine gestures via cheap privacy-preserving thermal sensors users train themselves; functional prototypes plus the user-defined service-selection architecture; next iteration was being lined up with Infineon (the company behind Google Soli's radar) and Skoda. Toolchain and enablement ran alongside: a personal-device integration bridge reused across seating buck tests, and hardware/software best-practice teaching inside the motion-sickness research team.

IMG-ICA-04 / gesture rig, training UI, on-device demo

IMG-ICA-06 / toolchain: device bridge and teaching rigs

The era's reference

Torsten Schonert, Director UX and Business Innovation at VW Group, on this era: within the first two weeks, board members were impressed by a working hardware-software prototype rather than slides.