Meghana is available for hire
Hire MeghanaMeghana Bhange
验证专家 in Engineering
软件开发人员
Location
加拿大蒙特利尔，QC
至今成员总数
2022年10月27日
Meghana is a machine learning engineer with a passion for solving problems in a data-driven manner. She is currently pursuing her Master's Degree with a research focus on privacy-preserving technologies at the Trustworthy Information Systems Lab, ÉTS Montréal. She has experience in natural language processing and has previously published work at SemEval-2020. Meghana is passionate about working on creative projects and always looks for new ways to apply her skills.
Portfolio
UInclude, Inc
自然语言处理(NLP)，机器学习，GPT...
ETS蒙特利尔的TISL实验室
Python 3，人工智能，Scikit-learn
自由端
人工智能(AI)、聊天机器人、OpenAI GPT-3 API...
Experience
Availability
Part-time
首选的环境
生成预训练变压器(GPT), 自然语言处理(NLP), AI Design
最神奇的...
...project I've worked on is developing an end-to-end custom recognition service on resource-constrained code-mixed settings with low latency requirements.
工作经验
AI Engineer
2023 - PRESENT
UInclude, Inc
- Developed a context-specific biased word matching model utilizing SpaCy and a Rule-based engine to identify biased words in job listings.
- Created a synonym enricher employing sentence transformer and GPT-3 to discover context-specific synonyms, 用无偏见的替代词取代工作列表中有偏见的词.
- Deployed the models using a FastAPI endpoint on AWS while storing and querying the data through DynamoDB.
Technologies: 自然语言处理(NLP)，机器学习，GPT, 生成预训练变压器(GPT), 亚马逊网络服务(AWS), Deep Learning, Python, TensorFlow, Scikit-learn, 亚马逊DynamoDB, Amazon S3 (AWS S3), FastAPI, SpaCy
Researcher
2022 - PRESENT
ETS蒙特利尔的TISL实验室
- Researched privacy-preserving ML and data publishing for a complaint redressal system.
- Researched model extraction attacks on machine learning systems with counterfactual explanation APIs.
- Modeled an adversary that can leverage the information provided by counterfactual explanations to build high-fidelity and high-accuracy model extraction attacks.
- 在Folktables数据集上对模型性能进行基准测试, 提取的模型保真度在97左右.6%.
技术:Python 3，人工智能(AI)， Scikit-learn
AI Developer
2023 - 2023
自由端
- Developed a FastAPl endpoint for a GPT-4 based chat interface tailored for parents and students. Successfully deployed the application on DigitalOcean, ensuring robust performance and scalability.
- 通过与LangChain集成增强了聊天端点, 整合像维基百科这样的插件, Search, and Math. 这种集成提高了信息的可靠性.
- Utilized a vector database to query documents for reliability of information retrieval. Developed endpoints capable of analyzing chat history to extract relevant topics and concepts.
Technologies: 人工智能(AI)、聊天机器人、OpenAI GPT-3 API, 生成预训练变压器(GPT), ChatGPT, LangChain
OpenAI开发人员
2023 - 2023
Zurney.app
- Built a FastAPI back end with GPT-3 API integration to generate a travel itinerary for a trip and extract locations. 然后用坐标对这些位置进行地理编码.
- Built a Next.js app to display the travel itinerary and show the geo-locations on Google Maps color-codes corresponding to days in the trip and information about each location.
- dockerization和部署FastAPI后端和Next.. js前端到DigitalOcean.
技术:人工智能(AI)、ChatGPT、OpenAI、Next.js, FastAPI, DigitalOcean, 大型语言模型(llm), 自然语言处理(NLP), 生成预训练变压器(GPT), GPT, Web开发, OpenAI GPT-3 API
机器学习工程师
2021 - 2022
Hunters.ai
- Researched and built analytical tools for evaluating threat-hunting detectors and understanding abnormal patterns in detection outputs.
- Organized the monitoring and quality check infrastructure in machine learning detectors.
- 创建了一个深入调查威胁的框架.
技术:Python 3, 机器学习, SQL, 人工智能(AI), Deep Learning, Scikit-learn, AI Design, 机器学习操作(MLOps), 亚马逊SageMaker, Kubernetes, APIs, 数据管道, 亚马逊网络服务(AWS), PostgreSQL, 工程数据, Amazon S3 (AWS S3), 亚马逊DynamoDB
机器学习工程师
2020 - 2021
The Verloop.io
- Contributed to the intent recognition service using a sentence transformer to improve the top-K recall and accuracy, 这将F1提高了40%.
- Designed, built, and deployed a multi-lingual name recognition service across all clients.
- Evaluated the performance of various language models like ULMFiT and VAMPIRE for low-resource language contexts.
- Created synthetic training data for FAQ systems in a chatbot using Generative Pre-trained Transformer 3 (GPT3) AI.
技术:机器学习, Python 3, Pandas, SpaCy, Chatbots, OpenAI, 人工智能(AI), Deep Learning, Scikit-learn, AI Design, Django, Google Cloud, 谷歌云平台(GCP), 语音识别, APIs, Web开发, 文本生成, 语言模型, 大型语言模型(llm), Flask, 数据管道, 机器学习操作(MLOps), 计算语言学, DaVinci, 生成预训练变压器(GPT), PostgreSQL
机器学习实习生
2019 - 2019
The Verloop.io
- 创建了为多语言对话定制的人名提取器. Tweaked Flair, Facebook的自然语言处理库, 用英语处理低延迟的用例, Spanish, and French.
- Improved the final model achieves by 47% in F1 compared to the previously deployed FastText mode.
- Deployed the developed multilingual name extractor to production with overall latency of under 500 milliseconds.
技术:人工智能(AI), Chatbots, Deep Learning, Scikit-learn, 文本生成, 语言模型, 大型语言模型(llm), Flask, PostgreSQL
Experience
使用反事实解释的模型提取攻击
A research project that I worked on with the Trustworthy Information Systems Lab at ETS Montreal. We researched how model adversaries can leverage the information provided by counterfactual explanations to build high-fidelity and high-accuracy model extraction attacks.
LitNER |文学命名实体识别
http://github.com/meghanabhange/litNER基于Spacy3在LitBank数据集上训练的命名实体识别. This project uses Roberta XLM as a base model and fine-tuned literature data to understand the terms generally used in literature. The pre-trained model released with the project can also be used to perform NER tasks on any literary text.
英语推特情感检测| SemEval2020
http://arxiv.org/abs/2008.09820这项工作增加了两种常见的方法, 微调大型变压器模型和采样高效方法，如ULMFiT. Prior work demonstrated the efficacy of classical ML methods for polarity detection. 我们对通用语言表示模型进行了微调, 比如伯特家族的人, which are benchmarked along with classical machine learning and ensemble methods. 我们发现NB-SVM比RoBERTa多6分.2%. The best-performing model is a majority-vote ensemble which achieves an F1 of 0.707.
维基百科教科书助手
http://github.com/meghanabhange/Wikipedia-Textbook-AssistantSimple streamlit application to be paired with textbooks so that one can easily extract the keyphrases from the text they are reading and get detailed information from Wikipedia to understand the relevant context.
Artificial Insanity (Cards Against Humanity with 稳定的扩散) | Toptal Hackathon
人工疯狂是一款面向Toptal黑客马拉松的多人游戏, 它使用稳定扩散生成文本提示的图像响应. In each round, 玩家选择提示, and other players generate a response using an integrated 稳定的扩散 interface.
I benchmarked performance in terms of quality and latency for DALLE and 稳定的扩散. Also, I deployed the final model on FastAPI to make it easier to integrate with the rest of the back end. 这个解决方案在黑客马拉松中获得了二等奖.
I benchmarked performance in terms of quality and latency for DALLE and 稳定的扩散. Also, I deployed the final model on FastAPI to make it easier to integrate with the rest of the back end. 这个解决方案在黑客马拉松中获得了二等奖.
Education
2023 - 2023
信息技术工程专业硕士学位(在读)
École de Technologie supsamrieure -蒙特利尔，加拿大
2016 - 2020
电子与通信工程专业本科以上学历
Savitribai浦那大学-印度浦那
Skills
库/ api
Pandas, Scikit-learn, SpaCy, TensorFlow
Tools
Slack，命名实体识别(NER)， 亚马逊SageMaker, ChatGPT
Frameworks
Django, Flask, Streamlit, Next.js
Languages
Python, SQL, Python 3
Storage
数据管道，PostgreSQL, Amazon S3 (AWS S3)， 亚马逊DynamoDB, Google Cloud
行业专业知识
Cybersecurity
Platforms
Kubernetes, 谷歌云平台(GCP), 亚马逊网络服务(AWS), Visual Studio Code (VS Code), DigitalOcean, AWS Lambda
Other
机器学习, 自然语言处理(NLP), 人工智能(AI), Deep Learning, APIs, 文本生成, 语言模型, GPT, 工程数据, Chatbots, OpenAI, AI Design, 机器学习操作(MLOps), 大型语言模型(llm), 计算语言学, 生成预训练变压器(GPT), OpenAI GPT-3 API, Research, 转移学习, BERT, Signals, 信息理论, Custom BERT, 稳定的扩散, DALL-E, FastAPI, Inference API, 语音识别, Web开发, DaVinci, Systems, Cryptography, 信息技术, 提示工程, LangChain
有效的合作
如何使用Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
1
分享你的需求
Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2
选择你的才能
Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3
开始你的无风险人才试验
与你选择的人才一起工作，试用最多两周. 只有当你决定雇佣他们时才付钱.
对顶尖人才的需求很大.
Start hiring