Professional Summary
WORK EXPERIENCE
Senior Software Engineer
Goava Sales Intelligence AB 🇸🇪
- Architected a core company data (financial reports, geographical location, corporate structure, industrial categories, URLs, situation, persons, etc.) & metadata indexing pipeline processing 2M+ company records daily, improving data retrieval speed by 28.6% and reducing query response time from 800ms to 570ms. This is 60% of the total system’s data (222M+).
- Implemented automated designation categorization system using GPT-4 API, reducing manual categorization effort by 40% and processing 10K+ designations weekly with 94% accuracy.
- Optimized AWS infrastructure costs by 43% through Elasticsearch query optimization and S3 access pattern improvements, reducing sync operation time by 86%.
- LLM based customer profile enrichment system, enhancing customer profiles with 15+ attributes, leading to a 12% increase in lead conversion rates.
- NER based company extraction & news tags detection tool with improving data extraction accuracy by 25% and reducing elasticsearch pressure by 38%.
Software Engineer
Goava Sales Intelligence AB 🇸🇪
- Developed an innovative company matching service for large file imports by users using AWS ElasticSearch, where users can perform the CRM sync operations.
- Participated in developing an event bus system in the Pub/Sub model to process the user request events.
- Developed a recruitment data pipeline to extract job postings from recruitment sites, such as Indeed.com.
- Built a keyword extraction tool and exposed it as an API for getting each company’s activity texts.
Junior Software Engineer
Goava Sales Intelligence AB 🇸🇪
- Developed collaborative filtering tool for finding similarities on some specific dimensions of data among the users’ set of companies’ data.
- Developed a dashboard and alert system using Grafana for the logs and metadata of several data pipelines. We integrated it into the Slack channel to notify us for any internal incidents.
- Done a couple of data modeling and schema design tasks for MySQL, Athena with Iceberg tables, and ElasticSearch.
SKILLS
- • Python
- • SQL
- • AWS ES
- • Scrapy
- • Playwright
- • PySpark
- • Kafka
- • Airflow
- • Pandas
- • Iceberg
- • Crawl4ai
- • Pydantic
- • Pytest
- • FastAPI
- • LLM
- • Langchain
- • Git
- • UV
- • Celery
- • LangGraph
CERTIFICATES
PROJECT
Scrapy, Celery, Playwright, Redis, Docker
- An effective hotel search web app which fetches the hotel info from booking.com and agoda.com based on user input and shows the best matches.
UV, Pydantic, Factory design pattern, SOLID principles
- An etl pipeline - fetch data from AWS Athena and store them into ES/OpenSearch & DynamoDB using UV package manager.
Pandas, Scikit-Learn, Ensemble, OneVsRestClassifier
- Vulnerability prediction for groundwater resources using DRASTIC model, which is a widely used method for assessing groundwater vulnerability.
EDUCATION
BSc in Computer Science and Engineering | 2015 - 2019
Shahjalal University of Science & TechnologyACHIEVEMENTS
- ACM ICPC Dhaka Regional 2018: I have participated with a team of three members. Rank: 63 among 298 teams.
- Hackathon: Champion of Hackathon in IUT 9th ICT Fest, 2017, Dept. of CSE, IUT.
- SQL(HackerRank): Solved all the problems and got five stars.
- Solved 1130+ problems in various online judges such as 215+ on Leetcode (136+ Medium), 135+ on Codeforces, 100+ on LightOj, etc.
v7.2.0
Professional Summary
WORK EXPERIENCE
Senior Software Engineer
Goava Sales Intelligence AB 🇸🇪
- Architected a core company data (financial reports, geographical location, corporate structure, industrial categories, URLs, situation, persons, etc.) & metadata indexing pipeline processing 2M+ company records daily, improving data retrieval speed by 28.6% and reducing query response time from 800ms to 570ms. This is 60% of the total system’s data (222M+).
- Implemented automated designation categorization system using GPT-4 API, reducing manual categorization effort by 40% and processing 10K+ designations weekly with 94% accuracy.
- Optimized AWS infrastructure costs by 43% through Elasticsearch query optimization and S3 access pattern improvements, reducing sync operation time by 86%.
- LLM based customer profile enrichment system, enhancing customer profiles with 15+ attributes, leading to a 12% increase in lead conversion rates.
- NER based company extraction & news tags detection tool with improving data extraction accuracy by 25% and reducing elasticsearch pressure by 38%.
Software Engineer
Goava Sales Intelligence AB 🇸🇪
- Developed an innovative company matching service for large file imports by users using AWS ElasticSearch, where users can perform the CRM sync operations.
- Participated in developing an event bus system in the Pub/Sub model to process the user request events.
- Developed a recruitment data pipeline to extract job postings from recruitment sites, such as Indeed.com.
- Built a keyword extraction tool and exposed it as an API for getting each company’s activity texts.
Junior Software Engineer
Goava Sales Intelligence AB 🇸🇪
- Developed collaborative filtering tool for finding similarities on some specific dimensions of data among the users’ set of companies’ data.
- Developed a dashboard and alert system using Grafana for the logs and metadata of several data pipelines. We integrated it into the Slack channel to notify us for any internal incidents.
- Done a couple of data modeling and schema design tasks for MySQL, Athena with Iceberg tables, and ElasticSearch.
SKILLS
- • Python
- • SQL
- • AWS ES
- • Scrapy
- • Playwright
- • PySpark
- • Kafka
- • Airflow
- • Pandas
- • Iceberg
- • Crawl4ai
- • Pydantic
- • Pytest
- • FastAPI
- • LLM
- • Langchain
- • Git
- • UV
- • Celery
- • LangGraph
CERTIFICATES
PROJECT
Scrapy, Celery, Playwright, Redis, Docker
- An effective hotel search web app which fetches the hotel info from booking.com and agoda.com based on user input and shows the best matches.
UV, Pydantic, Factory design pattern, SOLID principles
- An etl pipeline - fetch data from AWS Athena and store them into ES/OpenSearch & DynamoDB using UV package manager.
Pandas, Scikit-Learn, Ensemble, OneVsRestClassifier
- Vulnerability prediction for groundwater resources using DRASTIC model, which is a widely used method for assessing groundwater vulnerability.
EDUCATION
BSc in Computer Science and Engineering | 2015 - 2019
Shahjalal University of Science & TechnologyACHIEVEMENTS
- ACM ICPC Dhaka Regional 2018: I have participated with a team of three members. Rank: 63 among 298 teams.
- Hackathon: Champion of Hackathon in IUT 9th ICT Fest, 2017, Dept. of CSE, IUT.
- SQL(HackerRank): Solved all the problems and got five stars.
- Solved 1130+ problems in various online judges such as 215+ on Leetcode (136+ Medium), 135+ on Codeforces, 100+ on LightOj, etc.
v7.2.0