Adagrad AI Tech Stack

Computer vision algorithms for real-time camera deployments at scale

Software Development Pune, Maharastra 11–50 employees Founded 2018 Privately Held

Adagrad AI builds computer vision systems optimized for GPU and edge hardware—CUDA, TensorRT, NVIDIA Jetson, and Raspberry Pi dominate their stack. The project mix reveals a production-focused operation: highway traffic management, ATM solutions, CV algorithm optimization, and GenAI integration into vision pipelines. Post-sales friction (inventory risk, escalation delays, support inefficiencies) suggests they've crossed from R&D into field deployment and are now scaling operational maturity.

Tech Stack 21 technologies

Core StackC++ Python PyTorch scikit-learn pandas Linux C OpenMP CUDA GTest GStreamer TensorRT DeepStream C++17 Microsoft Office Excel NVIDIA Jetson Raspberry Pi

AdoptingGStreamer

What Adagrad AI Is Building

◆Challenges

Inventory stock-out risk
Issue escalation delays
Post-sales support inefficiencies

▲Active Projects

Highway traffic management system proposals
Solution architecture for atms
Cv algorithm development and optimization
Deployment of cv solutions on nvidia gpu infrastructure and jetson edge devices
Integrating genai into cv solutions
Post-sales lifecycle management of gateguard
Inventory coordination for deployment pipeline
Issue tracking and escalation system improvement

Hiring Activity

Minimal7 roles · 0 in 30d

Department

Engineering

Finance

Product

Sales

Seniority

Junior

Mid

Intern

Lead

Senior

Company intelligence

Find more companies like Adagrad AI by tech stack, pain points and active projects

Get started free

About Adagrad AI

Adagrad AI develops computer vision and deep learning solutions for real-world deployments across traffic management, financial infrastructure, and surveillance. Founded in 2018 by alumni from Microsoft Research, Georgia Tech, and similar labs, the company specializes in hardware-accelerated computer vision—running algorithms on cameras at scale using NVIDIA GPU infrastructure and edge devices like Jetson and Raspberry Pi. Their work spans algorithm development, model optimization, and post-deployment support. The team is based in Pune and operates at startup scale (11–50 people), with active hiring in engineering, product, and sales roles.

HeadquartersPune, Maharastra

Company Size11–50 employees

Founded2018

Hiring MarketsIndia

Frequently Asked Questions

What hardware does Adagrad AI deploy computer vision on?

NVIDIA Jetson, Raspberry Pi, NVIDIA DeepStream, and TensorRT for GPU acceleration. Deployments span edge devices and cloud GPU infrastructure.

What programming languages and frameworks does Adagrad AI use?

C++, C, CUDA, and Python. Core frameworks include PyTorch, TensorRT, and GStreamer for video streaming and hardware integration.