.png){kind=logo}
Srushti Manjunath
Mid-level Data Scientist specializing in NLP, LLMs, and cloud ML platforms
Connect with Srushti
Srushti 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/MLOps engineer who has shipped production systems for complaint intelligence and contact-center NLU, including LoRA/RLHF-tuned LLaMA models deployed on GKE with vLLM and Vertex AI batch pipelines to BigQuery. Demonstrates strong practical focus on hallucination control, data imbalance mitigation, and production monitoring (Langfuse) with regression testing and canary rollouts, plus experience orchestrating complex workflows with AWS Step Functions.
Experience
Education
Key Strengths
- Built and deployed LLM complaint-intelligence system (classification, summarization, compliance Q&A) to production
- Improved classification F1 by 15 points via LoRA + RLHF fine-tuning
- Achieved 88% faithfulness through grounding-focused RLHF and evaluator checks
- Production deployment on GKE with vLLM for low-latency inference; batch processing via Vertex AI to BigQuery
- Solved data imbalance using synthetic data + teacher-student distillation and curated rejection-sampled examples
- Strong MLOps/orchestration experience (Airflow, AWS Step Functions) with conditional branching and parallel workflows
- Designed auditable pipelines with monitoring/alerts (CloudWatch) and continuous drift monitoring (Langfuse) plus automated regression tests
- Effective cross-functional collaboration translating ops needs into measurable ML targets (F1, latency) and iterating with stakeholder feedback
Discover more candidates like Srushti
Search across thousands of pre-screened, high-quality, high-intent candidates on Reval.
Search Talent