Research Engineer
CurrentFine-tuning and training efficiency of open-source LLMDeveloping new methods for fine-tuningLeading evaluation efforts
Please complete the CAPTCHA to continue
A concise factual answer block for searchers comparing this professional profile.
George Grigorev is listed as Research Engineer @ Together AI at Together AI, based in London, England, United Kingdom. AeroLeads shows a matched LinkedIn profile for George Grigorev.
George Grigorev previously worked as Research Engineer at Together Ai and Senior Machine Learning Engineer at Snap Inc.. George Grigorev holds Степень Бакалавра, Программная Инженерия from Университет Итмо.
This section adds company-level context without repeating George Grigorev's masked contact details.
Review company-level records connected to George Grigorev before choosing the right outreach path.
As a machine learning engineer at Snap, I worked on cutting-edge generative AI projects that enhanced the user experience and engagement of the world's top chatbot, MyAI, and the popular Snap ML Lenses feature. I led and developed the My New Twin lens project, which achieved very high engagement metrics. Moreover in AI lenses project (Anime AI) I solved the problem of identity preservation in stylized images using diffusion models and my custom feature mapper. I also contributed to the improvement of intent classification, multimodality features, image captioning, and optimization of large language models (LLM) for MyAI.I have a strong background in software engineering, mathematics, and computer science, with bachelor's degrees from ITMO University and Saint Petersburg State University. I also have experience as a machine learning engineer at Reface, where I worked on text-to-speech, image and video retrieval applications, using technologies such as StyleGAN2, StyleCLIP, and CLIP. I am passionate about computer vision (CV), natural language processing (NLP), and generative AI, and I can easily implement state-of-the-art papers and scope big projects. I value innovation, collaboration, and impact, and I am always eager to learn new skills and technologies.
Company context helps verify the profile and gives searchers a useful next step.
A career timeline built from the work history available for this profile.
London Area, United Kingdom
Fine-tuning and training efficiency of open-source LLMDeveloping new methods for fine-tuningLeading evaluation efforts
London, England, United Kingdom
* Worked as part of MyAI team. This is top1 chatbot in the world according to coverage in the world. Prompt iteration for intent classification, adding multimodality features, improving image captioning, training multimodal LLMs, optimization of LLM inference.
London, England, United Kingdom
Previous successful projects:* Leading and developing of My New Twin lens project, which showed very high engagement metrics.(https://lens.snapchat.com/2897e0ee4e064728a6d259dca1a78f6c)* Solving the problem of identity preservation in stylized images. Come across with completely new pipeline of AI Lenses. Developed new type of lens experience based on.
Amsterdam, North Holland, Netherlands
- Text to Speech application * Research and development in Voice technologies (TTS) application, used in internal content creation studio. Comprises of text understanding and speech synthesis, as well as moderation of content and setting up Label Studio. * Created 3 different voices in the application in less than 6 months * Mentoring team of 3 people- New.
Moscow, Moscow City, Russia
* Development of Text Recognition system, including multilingual OCR and handwritten OCR in Russian language. Recognition quality increased from 65% to 75%, leading to massive cost reduction.* Incorporating Large Language Models for Post-OCR correction, including RoBERTa* Improving human in the loop system (verification by humans of low confidence samples.
Los Angeles, CA
* Speech Synthesis of celebrities. Developed ML model and backend from scratch using proprietary technologies and Amazon AWS infrastructure.* Background removal (Semantic segmentation) application, deployed on mobile devices to work in real time using ONNX and CoreML. Setup Toloka crowd-sourcing platform for labeling data and managed full process of.
Санкт-Петербург, Россия
* One of the first employees in the startup (only 3 people in R&D team)* Research and development in the Robotics sphere. Used such technologies as OpenCV, 3D Reconstruction, Reinforcement learning.* During my work company became producer of hardware robots andfound customers.
Санкт-Петербург, Россия
Image captioning task using Hadoop & Spark.Training of Large Language models using Transformer architecture
Санкт-Петербург, Россия
Consulting in the Data Science and ML.Notable projects: Anomaly detection from Tabular data. Classification of signals from Time-series data.
Санкт-Петербург, Россия
Churn rate prediction using Random Forest. Feature engineering and SQL.
Quick answers generated from the profile data available on this page.
George Grigorev works for Together AI.
George Grigorev is listed as Research Engineer @ Together AI at Together AI.
George Grigorev is based in London, England, United Kingdom while working with Together AI.
George Grigorev has worked for Together Ai, Snap Inc., Reface, Handl (Yc W20), and Matt.
You can use AeroLeads to view verified contact signals for George Grigorev at Together AI, including work email, phone, and LinkedIn data when available.
George Grigorev holds Степень Бакалавра, Программная Инженерия from Университет Итмо.
Search by job title, company, industry, location, and seniority. Export verified B2B contact data when you need it.
Start free trial Search contactsCheck these profiles if this is not the George Grigorev you were looking for.