Dinesh Jain
Applied Scientist 2 | Amazon | Generative AI | AWS | Microsoft R&D | Adtech | Search & Recommendation | NLP | LLAMA CPP | Reinforcement Learning | Computer Vision | Research
- Role
- Applied Scientist 2 at Amazon
- Location
- Hyderabad, TG, IN
- LinkedIn followers
- 500 followers
About Dinesh Jain
Experienced Applied Scientist with around 7 years of expertise leveraging AI/ML to solve large-scale business problems across advertising, finance, systems, and applied research domains.I hold a Master’s in Artificial Intelligence and Machine Learning from IIT Hyderabad and a B.E. in Electronics and Communication from NSIT Delhi.I’ve worked on a wide range of research and applied research projects across domains including Bayesian Inference, NLP, Computer Vision, Deep Generative Models, Causal Inference, Reinforcement Learning, Ads Ranking and Recommendation, Multimodal Learning, Multitask Learning, and LLM finetuning & optimization, Supplier Recommendation.Proficient in Python, C/C++, SQL, AWS, Git, and well-versed with modern ML/DL frameworks including TensorFlow, PyTorch, OpenAI Gym, TensorRT, ONNX, LLMs, and RAG pipelines.Always open to new challenges at the intersection of applied machine learning and cutting-edge ML research.
Experience
Applied Scientist 2
Sep 2024 — Present · Hyderabad, IN
Led development of an LLM-powered labelling and commodity recommendation system to minimize accounting defects in finance approvals and audits. Achieved Top-5 accuracy of 85.2% on catalog data, 81.4% on invoice data, and 75.6% on freeform data, driving higher classification reliability and improved audit efficiency. Finetuned MPNet embedding model on mix of human validated and LLM generated 2k samples which resulted in increase in Top-1 accuracy from 36.3% to 65.9%.Supplier recommendation identifying user query intent leveraging multiagent architecture along with multiple OpenSearch indexes for high precision. MRR currently stands at 0.42 without the exact supplier search and 0.47 including the exact supplier queries.AMLC(Amazon Machine Learning Conference) 2025 reviewer for GenAI categorySubmitted Research on commodity recommendation to Amazon CSS, 2026 ML track
Education
Netaji Subhas Institute of Technology
Bachelors, Electronics and Communications Engineering
2011 — 2015
Indian Institute of Technology Hyderabad
Masters, Artificial Intelligence & Machine Learning
2017 — 2019
Find verified contacts for anyone on LinkedIn
Unifers gives sales teams verified emails and direct dials, enriched profiles, and outreach that lands in the inbox.
Free plan included · No credit card required
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.