shivam@bengaluru● online
I build AI products that hold up in real workflows.
Product strategy, hands-on engineering, and applied AI, taking ideas from proof of concept to production. Most recently in AI hiring systems: assessments, screening, and conversational interviews.
13+years · technology & product
7+production AI systems
15+proofs of concept
$ ps --selected-work
Systems in production
idsystemwhat it doesstatus
001Enterprise Skills Assessment Platform
Coding and data-science assessments, source-code verification, remote proctoring, and analytics across 500+ skills.
assessments · code-verification · proctoring
AI-Assisted Candidate Screening
Embedding search and LLM ranking for JD–CV matching, question generation, and structured reports.
rag · vector-search · llm-ranking
Conversational Interviewing System
Agentic multi-turn interviews with dynamic follow-ups, live coding, and reviewer insights.
agents · conversational-ai · code-editor
$ ls ./lab --areas
15+ proofs of concept explored
résumé-intelligencesemantic-matchingrag-pipelinesagentic-workflowssource-code-evaluationvideo-analyticsspeech-and-response-evaluationconversational-ai
$ cat principles.md
How I work
01 · adoptionWorkflow fit before model cleverness
The AI that wins quietly fits into the flow of work and delivers outcomes consistently.
02 · validationProve it small, then productise
Validate as a PoC first; carry what holds up into production.
03 · platformPlatform judgement
Know when to build reusable capability and when to customise for an enterprise account.