Spaces:
Running
Running
File size: 2,232 Bytes
9c3c8b1 1c2029f 9c3c8b1 63239dd 1c2029f 63239dd 1c2029f 63239dd 1c2029f | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 | ---
title: README
emoji: ⚡
colorFrom: red
colorTo: indigo
sdk: static
pinned: false
---
# Codegnan IT Solutions
**Building and running self-hosted AI for Indian languages **
🌐 [Website](https://codegnan.com/) · 💼 [LinkedIn](linkedin.com/company/codegnan?originalSubdomain=in) · ✉️ [[CONTACT_EMAIL]](mailto:[hai@codegnan.com])
---
## About us
Codegnan IT Solutions builds, deploys, and operates AI systems on our own infrastructure,
using open-weight models. Our focus is practical AI for Indian users: Telugu and other Indic
languages, real-world audio, and production reliability over demos.
We also run AI training programs for students, professionals, and business teams.
## What we work on
| Area | Focus |
|------|-------|
| 🎙️ **Indic speech** | Speech-to-text for Telugu and other Indian languages, including noisy phone-call audio |
| 🗣️ **Voice agents** | Conversational AI agents that speak naturally in Indian languages |
| 📄 **Document intelligence** | OCR and document reconstruction for English and Indic scripts |
| 🧠 **LLM operations** | Serving, monitoring, and governing self-hosted open-weight LLMs |
| 💼 **AI hiring** | — resume matching, ATS analysis, quizzes, and AI interviews |
| 🎓 **Training** | Claude and AI bootcamps for business professionals and developers |
## How we work
- **Self-hosted first** — models run on our own GPUs; data stays in-house.
- **Open-weight models** — we credit and respect the licenses of the models we build on.
- **Indic-first** — every system is tested on Indian languages, not just English.
- **Measured, not assumed** — we publish numbers only when we have benchmarked them.
## What you'll find here
| Type | Status |
|------|--------|
| 🤖 **Models** | Coming soon |
| 📊 **Datasets** | Coming soon |
| 🚀 **Spaces** | Coming soon |
## For members
- Create repos **private by default**; make them public only after review.
- Never upload customer data, call recordings, API keys, or `.env` files.
- Every public model or dataset needs a card with its license and the base model it builds on.
- Naming: `[area]-[project]-[version]` — e.g. `speech-telugu-stt-v1`.
---
<sub>© Codegnan IT Solutions</sub> |