About: physical AI for spirits production.

Caskent is an independent startup building the autonomous operations layer for the world’s distilleries.

Mission

Encode distillers’ craft into bounded, auditable control loops that improve quality, yield, energy, and traceability without taking the palate out of command.

  • Perceive the run in real time.
  • Decide inside approved production limits.
  • Act through existing still, line, and warehouse systems.
VALUE LOOP
Outcome control loopA solid line chart shows variance narrowing as bounded autonomy engages. BASELINEAUTONOMY ENGAGEDHARD METRIC

Why now

Spirits producers face greater volatility in input costs, labor, energy, and quality expectations. Edge AI and sensing now make a closed loop feasible.

  • Demand

    Premiumization

    Consumers expect consistent character at scale.

  • Costs

    Input cost volatility

    Grain, energy, and cask costs make waste expensive.

  • Energy

    Decarbonization pressure

    Distillation remains heat intensive.

  • Labor

    Master-distiller shortage

    Critical process knowledge is scarce.

  • Technology

    Edge AI and sensing

    Real-time perception and control can run on site.

What we are building

Seven production agents

Ferment, cut, quality, maturation, blend, bottling, and energy control.

Digital twin

Simulation for distillation and maturation before the run is committed.

Assurance layer

Grounded outputs, audit logs, tenant isolation, and human checkpoints.

Principles

Trust before autonomy

Shadow and assist modes come before bounded control.

Compounding data moat

Each approved run improves site-specific and cross-process models while protecting recipes.

Land narrow, expand relentlessly

Start with a hard metric, then grow across the site.

Market opportunity

Market figures, forward-looking and used for company context.

$24BTAM
$6BSAM
$340M3-year SOM
~15%growth

Leadership

The founding team building Caskent.

LN

Laiba Nisar

Founder & CEO

Data scientist and full-stack AI developer pairing ML depth with business development to turn distilling data into scalable products.

LinkedIn
SR

Sharon Reddy

Co-founder & CTO

Data scientist with 4+ years building large-scale ML and generative-AI systems, from production RAG architectures to low-latency inference at scale.

LinkedIn

NVIDIA Inception and backing

Caskent has an NVIDIA Inception profile. We are exploring edge and simulation stacks such as Jetson, TensorRT, Triton, and Omniverse where they fit the distillery environment.

NVIDIA InceptionEdge AISimulationIndustrial autonomy

Roadmap

Forward-looking roadmap tied to design partners, paid conversion, and multi-site expansion.

  • 0-6 months

    Design partners and MVP

    Initial loops, edge runtime, and SOC 2 Type I in progress.

  • 6-12 months

    Paid conversion

    Graduated autonomy and SOC 2 Type II planning.

  • 1-3 years

    Full loop and multi-site

    OEM channel and scale toward $37M-$82M ARR.

  • 3-10 years

    Operations layer

    Autonomous operations for the world’s distilleries.

Careers

We are hiring builders across ML control, edge robotics, full-stack product, spirit science, and design-partner deployment.

View careers

Build with us.

Talk to Caskent about pilots, partnerships, or founding roles.

Contact us