Satish Sahu

Data Scientist

Bengaluru, Karnataka, India5 yrs 11 mos experience

Most Likely To SwitchAI Enabled

Key Highlights

6+ years of experience in Data Science and MLOps.
Expert in building production-grade NLP and machine learning solutions.
Proven track record of delivering business impact through scalable ML systems.

Stackforce AI infers this person is a Data Scientist specializing in SaaS and AI-driven solutions.

Contact

Skills

Core Skills

Machine LearningMulti-agent SystemsNatural Language ProcessingData ScienceComputer VisionOptical Character Recognition (ocr)Deep LearningWeb DevelopmentProgramming

Other Skills

AWS LambdaAWS CloudFormationAWS BatchLarge Language Models (LLM)JenkinsPrompt EngineeringMLflowDatabricks ProductsGo (Programming Language)Ragas evaluation for llmMetaBasen8nMCPStrands AgentSQL

About

Data Scientist with 6+ years of experience building production-grade NLP, machine learning, and AI solutions. Expert in recommendation systems, nationality prediction, OCR, search algorithms, demand forecasting, customer segmentation, affinity, and propensity modeling across diverse domains. **Production MLOps Expertise**: - End-to-end ML pipelines: MLflow, Kubeflow - Model deployment: Docker, Kubernetes, AWS SageMaker, AWS Batch, ECS,ECR - CI/CD for ML: GitHub Actions, Jenkins - Monitoring & retraining: Grafana, MLFlow - Scalable inference: FastAPI,Flask Proven at delivering **business impact** through robust, scalable ML systems.

Experience

5 yrs 11 mos

Total Experience

11 mos

Average Tenure

1 yr

Current Experience

Netcore cloud

Assistant Manager - Data Scientist

May 2025 – Present · 1 yr · Bengaluru · Hybrid

AI for Marketing

Multi-agent SystemsAWS LambdaAWS CloudFormationAWS BatchLarge Language Models (LLM)Jenkins+10

Foundit

Data Scientist

Mar 2024 – Mar 2025 · 1 yr · Bengaluru · Hybrid

◦ Realtime Profile Summary Update Pipeline: Developed a real-time pipeline to dynamically update candidate
profile summaries using prompt engineering, SQL,OpenAI, and RabbitMQ. Integrated LLM-based summarization
techniques with event-driven architecture, ensuring up-to-date and contextually relevant profiles. This enhanced
candidate visibility and improved recruiter engagement by 20%.
◦ Regular JR and Similar Job JR: Built a job recommendation system to match users with relevant jobs based on
their profile (Regular JR) and jobs they previously applied to (Similar Job JR). Explored all-MiniLM-L6-v2 and
job-BERT, and finally implemented using OpenAI embeddings,elastic search. This system increased the job
application rate by 100%.
◦ Nationality Prediction: Developed a nationality prediction model for 2.7 million Singaporean users,
leveraging LSTM-based name classification and rules-based logic. Achieved 88% precision and 87% recall using
PyTorch, Python, and FastAPI, significantly improving prediction accuracy.
◦ Contextual Synonyms: Engineered a contextual synonym generator for designations using Elasticsearch and OpenAI
with prompt fine-tuning. Implemented fuzzy matching (0.9 threshold), acronym detection from a master dictionary,
and functional role-based synonyms for improved search relevance.
◦ Industry Derivation: Built an industry derivation system for 0.8 million companies, aligning with Crunchbase
standards. Integrated Python, Elasticsearch, and OpenAI to enhance classification accuracy, improving the
experience for 64 million users.
◦ AI-Generated Job Descriptions: Designed an AI-powered job description generator for passive job roles using
OpenAI and prompt fine-tuning. Leveraged recruiter emails and subjects to craft highly relevant job descriptions,
effectively engaging over 2 million active customers.

SQLOpenAIRabbitMQElasticsearchPyTorchPython+3

Betterplace

Senior Data Scientist

Sep 2023 – Mar 2024 · 6 mos · Remote

In-House id card OCR: Developed an in-house aadhaar ocr using AWS textract,FastApi,MongoDB which assist clients onboarding frontline workers in India,Indonesia and Saudi Arabia to fetch their basic details from documents their ID cards and documents within milliseconds.Reduced cost of onboarding using OCR by 50% for international clients and national clients. Worked on Blur detection, Rotation detection, Id-Card Authentication,Orientation detection to enhance the overall Field Extraction Rate of OCR and increased the closure rate to 90 %.
Models Explored : CNN,VGG16,Xgboost,Squeezenet,Doctr,PaddleOCR
◦ Q&A bot using LLM POC: Built an QA bot using fastapi ,mongoDB,sentence transformer,openAI Api for clients to enhance their customer service. Used mongo db for caching frequently asked questions to save and enhance productivity of the Q &A bot.

AWS TextractFastAPIMongoDBComputer VisionOptical Character Recognition (OCR)Python+3

Reliance retail

Data Scientist - Manager

May 2022 – Jul 2023 · 1 yr 2 mos · Gurugram, Haryana, India

Key Achievements:
> Developed a state of the art product search algorithms in less than 6 weeks using DSA, Data Science
and SQL, which assist million users to fetch their desired products within milliseconds, saves 60% of a
customer's time to find and add a product to their basket and 20% search accuracy than the older logic.
> Worked on customer segmentation to cluster
user into different clusters using Machine helping the respective marketing team to target customers based on pattern driven clusters.
>Designed Product Variant Catalog using nlp
>worked on Demand Forecasting and inventory management Time series model using Prophet and TFT.
Key Responsibility:
> Driving the digital transformation at MilkBasket. Helping the organisation to grow through data insights
and AI/ML.

Data ScienceSQLMachine LearningProphetDockerPython

Tata consultancy services

Data Scientist

Jun 2020 – May 2022 · 1 yr 11 mos · Bengaluru, Karnataka, India

Worked on various projects related to Artificial Intelligence.
Familiar with Machine learning and Deep learning algorithms.
Having a good Airlines Domain knowledge.
Worked on airlines data for a world renowned airlines company.
Developed a state of the art predictive model from scratch for the prediction of ETA for world renowned airline's user using xgboost from scratch.
Tech Stack: Python, Machine Learning, Deep Learning,Flask,Postman,Keras

Machine LearningDeep LearningPythonFlaskPostmanKeras

Dichroic labs llp

Aws

Jun 2020 – Jun 2020 · 0 mo · Mumbai, Maharashtra, India

Hosted a dynamic web page.Created the architecture for hosting website which includes VPC,Internet gateway,Public Private Subnets,Aws Rds,DynamoDb etc. Created a coding platform for developers using code commit,code pipeline and code deploy in which code privacy was provided using IAM.