AI for HR tech in India

HR teams and HR-tech products in India draft an enormous amount of repetitive text — job descriptions, interview questions, policy answers, performance-review language — and a language model can take the first pass on most of it. unoblox gives HR and people teams one rupee-billed, OpenAI-compatible endpoint for GPT, Claude, DeepSeek, Qwen and more, with a single monthly GST invoice instead of a foreign-currency subscription per tool.

Where HR teams actually plug this in

The practical uses are drafting and summarising: writing a first draft of a job description from a few bullet points, summarising a resume against a role's requirements for a recruiter to review, generating candidate-neutral interview questions for a panel, answering common HR-policy questions in a chatbot, and helping a manager phrase performance-review feedback more clearly. A recruiter or manager makes every actual decision — unoblox has no scoring or shortlisting logic built in.

Choosing a model at HR volumes

Model	₹ / 1M input	₹ / 1M output	Typical use
qwen/qwen3-1.7b	0 (free)	0 (free)	Free pilot for a policy FAQ bot
deepseek-ai/deepseek-v4-flash	9.07	18.14	High-volume resume summarisation
google/gemma-4-31b-it	13.10	38.30	JD drafting, interview-question generation
openai/gpt-5-mini	25.2	201.6	Production-quality policy chatbot

More OpenAI, Anthropic and open-weight models are on the platform with live pricing at /models.

Multilingual employee communication

India's workforce often needs HR communication in more than English and Hindi — a policy summary might need to go out in Tamil, Bengali or Marathi. Use a general multilingual model such as qwen/qwen3-235b-a22b-instruct-2507 or meta-llama/llama-4-scout-17b-16e-instruct for the first draft, and route it through someone fluent in that language before it's sent company-wide.

A sample call

curl https://api.unoblox.ai/v1/chat/completions \
  -H "Authorization: Bearer ub-gw-xxxxxxxxxxxxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "google/gemma-4-31b-it",
    "messages": [{"role": "user", "content": "Draft a job description for a backend engineer from these five bullet points."}]
  }'

Being careful about what this shouldn't do

A resume summary or a suggested interview question is a starting point, not a hiring decision, and unoblox does not claim any bias-testing or compliance certification for HR use — that responsibility stays with your process and your legal team. On data: only the free qwen/qwen3-1.7b model runs on infrastructure unoblox controls; other models run on their original providers' infrastructure outside India, so treat candidate PII accordingly.

Frequently asked questions

Can this screen or shortlist candidates automatically? No. It can summarise a resume against a role for a recruiter to review, but the decision stays with your team.

Is candidate data stored in India? Only for the free qwen/qwen3-1.7b model. Other models run on their original providers' infrastructure outside India.

Which model should we use for regional-language employee communication? A general multilingual model such as Qwen3 235B-A22B or Llama 4 Scout, reviewed by a fluent speaker before company-wide use.

Is there a free tier to try before we commit budget? Yes, qwen/qwen3-1.7b is free to call for prototyping.

Can we switch models later without redoing our integration? Yes, every model sits behind the same base URL and key, so switching is a one-line change.

How is this billed for an HR software budget? In rupees, on one monthly GST invoice with input tax credit claimable, covering every model used.

Get started in rupees → https://unoblox.ai/sign-in