Prototype → MVP → Scale
Neev Solutions builds prototypes, MVPs and production systems across Industrial IoT, Computer Vision, Fintech, Food-service and private AI.
One engineer’s span, from the sensor to the screen. Firmware, services, cloud and AI — designed together, so nothing gets lost at the hand-offs. Deterministic where it must be. Auditable everywhere. Built to still be running in five years.
Solutions
01 / 05
From the sensor on the factory floor to the dashboard in the boardroom.
Edge gateways, telemetry pipelines and multi-tenant monitoring platforms for industrial and process data. Built to run unattended for years, not weeks.
02 / 05
Cameras that understand what they see — on the edge, offline.
Visual inspection, counting, OCR and document-layout understanding. Models are evaluated on your data and deployed where the camera is, with a human-review loop where it matters.
03 / 05
Money movement with the audit trail regulators expect.
Payment flows, pricing engines and ledgers with strong identity built in — passkeys, OAuth2 + PKCE, idempotent APIs and tenant-isolated data by default.
04 / 05
Order, pay, pick up. No queue, no paper.
Employee canteen platforms and seat-side ordering for venues: mobile apps, kitchen displays, receipt printing and admin consoles that operators actually enjoy using.
05 / 05
AI you can audit, running on hardware you control.
Retrieval-augmented systems, document comparison and compliance reasoning on local LLMs. Deterministic rules decide; language models explain — every answer carries its citation.
Selected work
Offline comparison of engineering standards and policies: what changed between editions, how much it matters, and exactly where it’s written.
Theatre guests scan the QR on their seat, order for the interval, and collect in seconds. Kitchen works from a live board during act one.
Mobile ordering, UPI payments and kitchen printing for a corporate cafeteria: employee app, staff app and admin console on one backend.
Engineering experience on an established multi-tenant industrial reporting platform, from backend services to cloud infrastructure.
An offline, OpenAI-compatible LLM service that installs on a Linux or macOS box in one step and heals itself.
AI research agents investigate an investment theme, every claim is checked against its sources, and humans approve each step.
A mobile proof of concept that turns lab reports into a verified, searchable health timeline the patient owns and shares on their own terms.
Live lab
Three miniature versions of systems we build. Everything runs in your browser — nothing you type leaves this page.
Press Analyse change to compare the two editions.
Simulated sensor at 20 Hz with a self-learning normal band. Anomaly alerting of the kind that runs on edge gateways.
Simulated detector output on a conveyor feed. Counting and threshold tuning as an operator would see it.
How we engage
01
2–4 weeks
Prove the risky part first. A working slice that answers one hard question — will the sensor data hold up, can the model read this document, will users pay this way.
02
6–12 weeks
A product real users can rely on. Auth, payments, CI/CD and monitoring included — not bolted on later.
03
Ongoing
Take what works to hundreds of customers. Kubernetes, infrastructure as code and the boring reliability work that keeps uptime at three nines.
Toolbox
No lock-in. Every system we deliver can be run, inspected and extended by your own team.
Edge & Devices
Services
Data
AI
Platform
Identity
Next step
Tell us what you’re building. You’ll get an honest answer within two working days — including when we’re not the right fit.