Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I am a Machine Learning Engineer with experience building and deploying AI solutions across computer vision, multimodal learning, and generative AI. At IBM, I delivered 17+ Projects, including a document intelligence bot, a cybersecurity risk advisor, and a Java code assistant that boosted developer productivity by 30%. Previously, I worked on large-scale image retrieval systems for criminal forensics and developed OCR pipelines, insurance claim automation agents, and multimodal vision-language models. My strength lies in taking AI from research to production, designing end-to-end systems that solve real-world problems at scale.…I am a Machine Learning Engineer with experience building and deploying AI solutions across computer vision, multimodal learning, and generative AI. At IBM, I delivered 17+ Projects, including a document intelligence bot, a cybersecurity risk advisor, and a Java code assistant that boosted developer productivity by 30%. Previously, I worked on large-scale image retrieval systems for criminal forensics and developed OCR pipelines, insurance claim automation agents, and multimodal vision-language models. My strength lies in taking AI from research to production, designing end-to-end systems that solve real-world problems at scale.

Sanyam Suri

AI Engineer, Data Scientist, AI Developer





I am a Machine Learning Engineer with experience building and deploying AI solutions across computer vision, multimodal learning, and generative AI. At IBM, I delivered 17+ Projects, including a document intelligence bot, a cybersecurity risk advisor, and a Java code assistant that boosted developer productivity by 30%. Previously, I worked on large-scale image retrieval systems for criminal forensics and developed OCR pipelines, insurance claim automation agents, and multimodal vision-language models. My strength lies in taking AI from research to production, designing end-to-end systems that solve real-world problems at scale.…I am a Machine Learning Engineer with experience building and deploying AI solutions across computer vision, multimodal learning, and generative AI. At IBM, I delivered 17+ Projects, including a document intelligence bot, a cybersecurity risk advisor, and a Java code assistant that boosted developer productivity by 30%. Previously, I worked on large-scale image retrieval systems for criminal forensics and developed OCR pipelines, insurance claim automation agents, and multimodal vision-language models. My strength lies in taking AI from research to production, designing end-to-end systems that solve real-world problems at scale.

Available to hire

I am a Machine Learning Engineer with experience building and deploying AI solutions across computer vision, multimodal learning, and generative AI. At IBM, I delivered 17+ Projects, including a document intelligence bot, a cybersecurity risk advisor, and a Java code assistant that boosted developer productivity by 30%. Previously, I worked on large-scale image retrieval systems for criminal forensics and developed OCR pipelines, insurance claim automation agents, and multimodal vision-language models. My strength lies in taking AI from research to production, designing end-to-end systems that solve real-world problems at scale.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Language

English

Fluent

Hindi

Fluent

Work Experience

AI Engineer at IBM

October 1, 2023 - Present

● Developed a cybersecurity risk advisor platform using ElasticSearch VectorDB to query thousands of documents and generate LLM-driven risk scores and remediation. Resulted in saving 0.25–2 hours per vulnerability per analyst. ●Built a Document Intelligence Bot for a banking client to summarize, translate, and answer queries on 20+ page circulars using LLaMA3-70B (q) and LangChain, deployed on-prem in an air-gapped environment. ● Led development of a Java Code Assistant using to auto-generate backend microservices, boosting developer productivity by 30%. ● Delivered 17+ GenAI Projects (team-high) across enterprise domains using Huggingface, PyTorch, LangChain, CrewAI, RAG, VectorDBs and prompt engineering—enabling rapid prototyping and validation of LLM use cases.

Research Engineer at Staqu Technologies

June 1, 2022 - October 1, 2023

● Built and trained text-based image retrieval system using BLIP and Vision Transformers for criminal forensics, delivering 91% accuracy and 1s retrieval time on a 5K image gallery in production deployment. ● Improved action recognition pipeline by replacing legacy R(2+1)D models with newer SOTA architectures (e.g., STTS), by finetuning and boosting accuracy by 15% and tripling inference speed. ● Collaborated on person re-identification research with IIT Delhi, implementing and enhancing the SOTA PASS model with custom training pipelines across supervised, UDA, and unsupervised methods. ●Finetuned real-time CV/ML models with >90% accuracy and 500+ FPS across detection, classification, and retrieval.