The problem
Recruiters were spending most of their week on the least differentiated part of the job — reading CVs, chasing candidates for salary and notice details, and running repetitive screening calls.
Engagement detail
- CLIENT
- Talent acquisition team
- INDUSTRY
- Human resources
- DISCIPLINE
- Agentic AI
PythonLangChainGPTElevenLabsWhatsApp APIFlaskChroma DB
What we built
- Built bulk CV and job-description ingestion with LLM-based semantic scoring, ranking candidates against the specific role rather than keyword overlap.
- Integrated the WhatsApp Business API to collect current CTC, expected CTC and current employer conversationally, at the candidate’s convenience.
- Created a voice-based first-round interview module using ElevenLabs for realistic spoken question-and-answer, with responses transcribed and scored by GPT against the role criteria.
- Designed a candidate scoring and tracking pipeline that leaves the organisation with a reusable, searchable talent database.
Available for new engagements
Have a problem shaped like this one?
We will tell you what transfers from this engagement to yours, what does not, and what we would do differently now.