About Scribara

Give clinicianstheir time back.

Scribara exists to become the cognitive layer between physician and patient — starting by finishing the documentation that steals two hours of every clinical hour. We build owned clinical AI models on NVIDIA compute, not wrappers around someone else's API.

Join us Contact

Specialty-first · owned models · NVIDIA Inception

Epic FHIRathenahealthNVIDIA InceptionHL7SOC 2HIPAA

Mission

Why we started

Specialty physicians spend roughly two hours documenting for every hour of care, and today's ambient scribes stop at the note — leaving coding, prior auth, referrals, and follow-up to overstretched staff. We believe that entire layer of administrative cognition can be done well by AI that is specialty-first and owns its models.

Scribara is not a ChatGPT wrapper. It is a vertically integrated system of record and action, built on owned clinical models running on NVIDIA compute, with a data flywheel that compounds with every encounter.

Own the intelligence.

Renting models caps the ceiling. We train and serve our own — speech, reasoning, and vision — on accelerated compute.

Trust is the product.

In medicine, an unreliable output erodes everything. We invest disproportionately in verifiers, evals, and human-in-the-loop.

Specialty-first.

Depth beats breadth. We win one specialty's workflow completely before the next.

Elevate, don't fabricate.

We publish what we can prove and label what we can't — aspirational and unknown, never dressed up as fact.

Company

Scribara today

Specialties at launch

Products in the ecosystem

Team members

Target flywheel encounters [ASPIRATIONAL]

Leadership

Co-founders

Two builders who believe AI should own its intelligence, not rent it.

Sabir is an AI/ML engineer and certified Data Scientist who has built intelligent automation systems and agent-based AI from the ground up. At Tkrupt he designed voice-interactive AI agents using VAPI and optimized workflows through n8n automation. His work at Scribara is focused on the agentic encounter-completion loop — the system that finishes the clinical encounter from note to prior auth. Sabir holds a BS in Computer and Information Sciences from PIEAS.

AI Agents
LangChain
NLP
AI Automation
Siamese GNN
Python
n8n

BS Computer & Information Sciences · PIEAS (2021–2025)

Hibah transforms raw datasets into actionable intelligence. At The Alpha's Digital she built dashboards, automated reporting pipelines, and drove data-driven decision-making using Python, SQL, and KQL across cross-functional teams. At Scribara she owns the data infrastructure — the pipelines, dashboards, and analytics that power the data infrastructure and Nexar population intelligence layer. Hibah holds a BS in Computer Science from the University of Lahore.

Python
SQL
KQL
Power BI
Data Pipelines
Statistical Analysis
EDA

BS Computer Science · University of Lahore (2020–2024)

Engineering

Key engineers

Deep AI and data expertise across NVIDIA compute, MLOps, machine learning, and analytics.

Omar has built production AI audio pipelines with Triton Inference Server at <10% error rate, optimized LLM inference via TensorRT-LLM on H100 GPUs, and deployed MLOps workflows on Azure ML. He holds NVIDIA DLI certificates in CUDA Python and RAPIDS. At Scribara he leads the accelerated-compute serving layer — TensorRT-LLM, Triton, and NIM packaging.

TensorRT-LLM
Triton
RAPIDS
CUDA Python
MLOps
Azure ML
Docker

MS Robotics & Intelligent Machines · NUST (2017–2021)

Nazia specialises in deep learning and computer vision. She developed signature-forgery detection systems using DenseNet121 and VGG16 architectures, and holds Google Data Analytics and Supervised ML certifications. At Scribara she focuses on the medical imaging and computer vision pipeline — specialty medical-imaging CV for dermatology, ophthalmology, and orthopedics on Clara/MONAI.

TensorFlow
PyTorch
Computer Vision
Scikit-learn
DenseNet / VGG
Python
Power BI

MSc Data Science · Mehran UET (2025) · BE Computer Sys Eng

Areesha has nearly three years of progressive data experience at Telerelation, working from intern to Associate Data Analyst while simultaneously running independent data science projects. She applies Python, SQL, Power BI, and Tableau to surface operational intelligence from complex datasets. At Scribara she builds the analytics and reporting infrastructure — the dashboards and data pipelines that surface quality insights for the Nexar and Revix layers.

Power BI
Python
SQL
Tableau
Data Visualization
Analytics

BS Computer Science · NUST (2020–2024)

How we work

Operating principles

A small, focused team that out-ships incumbents.

Forward-deployed

Founders embed with design partners — we stay in the exam room until the workflow is right.

Eval-driven

Golden datasets gate every release. We measure accuracy, not features shipped.

Platform mindset

Build the layer, not the feature. Every line of code should compound the moat.

Open where it helps

SDKs, eval frameworks, and benchmarks build trust with developers and buyers.

Write things down

Clarity compounds like the flywheel. We document decisions, not just code.

Outcome-obsessed

We measure hours returned to clinicians, not velocity points or press mentions.

Proof

What clinicians say

Outcomes are reported by design partners and marked illustrative [PLACEHOLDER].

"The cognitive layer between physician and patient — that's the ten-year bet."

Founding team · Scribara [PLACEHOLDER]

"We measure success in evenings returned to clinicians."

Founding team · Scribara [PLACEHOLDER]

Trust

Enterprise-grade from day one

Healthcare buys on trust. Scribara is built to clear procurement, security review, and compliance.

HIPAA / BAA

PHI handled under a Business Associate Agreement.

SOC 2 Type II

[ASPIRATIONAL] — roadmap to Type II within 12 months.

On-prem NIM

Run owned models inside your datacenter via NVIDIA AI Enterprise. [ASPIRATIONAL]

Immutable audit

Every agent action logged, reversible, tied to the clinician's signature.

Data residency

US, EU, UK regions; per-tenant keys (BYOK option). [ASPIRATIONAL]

ISO 42001

AI management-system conformity on the roadmap. [ASPIRATIONAL]

Answers

Frequently asked

Where are you based?

The founding team is based in Pakistan — Islamabad and Lahore. We are building for the global specialty medicine market.

Are you hiring?

Yes — see Careers for open roles across AI/ML engineering, clinical NLP, MLOps, and clinical domain expertise.

Who backs Scribara?

Scribara Labs has raised a $2M seed round led by Raed Ventures. We are also aligned with NVIDIA Inception for compute, ecosystem access, and co-marketing in the healthcare AI space.

Help build the cognitive layer.

Join a team returning evenings to clinicians — and building AI that owns its intelligence.

See open roles Contact us