.png){kind=logo}
Drashti Patel
Junior Software Engineer and ML Researcher specializing in full-stack and applied deep learning
Connect with Drashti
Drashti already has a relationship with Reval, so a warm intro from us gets a much better response than cold outreach.
Recommended
Already have an account?
About
LLM engineer who built a production-style educational questionnaire generation system (MCQs/fill-in-the-blanks/short answers) using Hugging Face models (BERT/T5) and implemented grounding, decoding tuning, and post-generation validation to control hallucinations and quality. Also developed a "tech care" assistant chatbot with a custom Python orchestration/router layer (intent classification, context management, multi-step flows) and a structured testing/evaluation approach including expert review and automated checks.
Experience
Education
Key Strengths
- Built and evaluated an LLM-based educational question generation system with grounding and quality controls
- Designed multi-layer evaluation: gold dataset benchmarking + automated validation rules + expert manual review
- Strong practical approach to reducing hallucinations via prompt constraints and post-generation filtering
- Custom-built lightweight orchestration layer (intent routing, context management, multi-step flows) similar to LangChain/Haystack concepts
- Systematic agent/workflow engineering with modular design and unit/integration/E2E test coverage
- Effective collaboration with non-technical stakeholders through early prototypes and plain-language requirement translation
Discover more candidates like Drashti
Search across thousands of pre-screened, high-quality, high-intent candidates on Reval.
Search Talent