Vahan.ai Supercharges AI Hiring With NVIDIA Nemotron to Target India’s Multilingual Workforce

Vahan.ai is taking its AI recruitment platform deeper into India’s linguistic diversity, fine-tuning NVIDIA Nemotron 3 Nano to deliver faster, cheaper, and more human-like hiring conversations for blue-collar workers.

India’s blue-collar hiring market is getting an AI upgrade, and Vahan.ai bets the next breakthrough will come from models built specifically for the country’s workforce.

The AI-powered recruitment company has fine-tuned the open NVIDIA Nemotron 3 Nano model for its multilingual AI recruiter, working with NVIDIA technical experts through the NVIDIA Inception program. The move is designed to improve recruitment conversations while dramatically reducing response times and inference costs.

The development could become particularly significant as Vahan.ai expands beyond English and Hindi toward eight additional Indian languages.

Vahan.ai Says Its AI Recruiter Is Getting Faster

Vahan.ai’s voice-based AI recruiter lets job seekers discover opportunities, apply for positions, and schedule interviews through ordinary phone conversations.

That approach targets one of India’s biggest employment challenges: candidates with limited digital literacy, inconsistent internet access, or difficulty navigating recruitment platforms in English.

The company says its platform has already facilitated more than 1.5 million job placements across over 920 Indian cities and connected nearly 50 million Indians with employment opportunities.

The new NVIDIA Nemotron-powered system aims to make those conversations more efficient at scale.

According to Vahan.ai, benchmarking across seven recruitment-focused parameters showed the fine-tuned model outperforming both the original Nemotron model and the company’s previous cloud-hosted 120-billion-parameter model.

6.7x Faster Responses Could Change the Hiring Experience

The headline performance numbers are striking.

Vahan.ai says its optimized deployment delivered almost 6.7x faster time-to-first response and more than 3x lower average end-to-end response latency.

The company also says inference costs dropped substantially.

For a recruitment platform handling thousands of simultaneous conversations, those improvements could directly affect how quickly candidates move through the hiring funnel, particularly in high-volume sectors such as food delivery, quick commerce, mobility, warehousing and manufacturing.

Instead of simply making a general-purpose AI model smaller or faster, Vahan.ai is tailoring the technology to the specific conversations and workflows involved in frontline recruitment.

Why India’s Languages Are a Huge AI Challenge

Recruitment conversations in India are rarely as straightforward as selecting a language from a dropdown menu.

Candidates may switch between languages during a conversation, use regional expressions or communicate in ways shaped by local employment conditions.

Vahan.ai trained the Nemotron model using proprietary production datasets containing human-corrected multilingual recruitment conversations. The goal was to teach the model not just to understand Indian languages but also to handle recruitment-specific tasks and function execution.

The company says the model maintained stronger performance even when tested on previously unseen recruitment conversations.

That ability to generalize could be crucial as Vahan.ai expands its language coverage.

The Bigger Bet: Small Language Models for Hiring

The Nemotron deployment is being positioned as more than a single technology upgrade.

Vahan.ai says it is the first step toward developing specialized Small Language Models (SLMs) designed specifically for India’s blue-collar recruitment ecosystem.

The company has also established a continuous evaluation and fine-tuning pipeline that can use production conversations to improve the system over time.

Founder and CEO Madhav Krishna said the company’s objective is to build AI specifically for recruitment rather than depending entirely on general-purpose models.

If that strategy succeeds, Vahan.ai could offer a blueprint for specialized AI systems designed around India’s workforce rather than simply adapting global models to local markets.

With millions of workers and employers already using its platform, the company is now trying to make the recruitment process not only more automated, but faster, cheaper, and more accessible across India’s linguistic landscape.