Sanjeet Attili
Data Scientist
Bengaluru,KA
Summary
Data Scientist with over 4 years of experience in developing innovative machine learning solutions. Expertise encompasses Representation Learning, Semantic Matching Systems, Named Entity Recognition, text classification, and model performance optimization.
Overview
2
2
Certifications
4
4
years of professional experience
Work History
MLE-II
Interview Kickstart
Bengaluru
07.2025 - Current
- Spearheaded the creation of an automated mock interview matching system that pairs students with the most suitable interviewers. This solution enabled Student Success Advocates (SSAs) to move away from manually scouring fragmented, unorganized Excel data, reducing the time required to identify ideal profiles from approximately 30 minutes to nearly instantaneous results.
- Developed an AI Analytics agent that automates Metabase dashboard creation from BigQuery tables using PRD-style inputs. This initiative streamlined the data analysis process for Product Managers and business stakeholders, cutting down manual effort from several hours to just minutes.
- Developed an automated pipeline for weekly business call analysis leveraging GPT-5-mini and HDBSCAN, significantly improving reporting workflows and providing actionable insights on course performance across business units. The output is actively used by department heads during their weekly business review meetings to get insights on market trends from the conversations of the leads.
- Boosted lead conversion rates by 23% compared to previous marketing heuristics by developing custom features and training XGBoost models on Indian market datasets with over 100k samples.
- Pioneered adoption of advanced AI development tools early on (e.g., Cursor/Claude) to build custom skills for code streamlining and integrate optimizations (e.g., MCP's), driving personal coding efficiency and sharing key AI advancements with the team regularly.
- Served as an active member of the MLE hiring panel, evaluating technical ML talent.
Data Scientist
Meesho
Bengaluru
06.2024 - 07.2025
- Optimized core matching services for 150M+ active products by building and maintaining FAISS-based similarity search solutions, ensuring high performance and reliability.
- Enhanced operational efficiency and reduced Databricks-related overhead by successfully migrating and streamlining batch job pipelines into internal Airflow and Prism environments.
- Drove matching quality improvements by experimenting with BEIT2 models on a 12M+ dataset, achieving a 1-2 percentage point lift in precision and recall while reducing false image pairs.
Associate Data Scientist
Uniphore
Bengaluru
07.2022 - 06.2024
- Developed Docker-based service to automate data annotation for text classification and Named Entity Recognition (NER), leveraging Large Language Models like Llama2 (7B/13B) and UniversalNER-7B with a multi-agent debate prompting strategy to minimize manual entity labeling effort.
- Worked on Docker-based service to automate data annotation for text classification and Named Entity Recognition (NER). By leveraging Large Language Models such as Llama2 (7B/13B) and UniversalNER-7B alongside a multi-agent debate prompting strategy, in view of minimizing the manual effort required for entity labeling.
- Reduced the latency of a T5-based Setfit model by exporting it to ONNX and performing graph optimizations.
- Worked on Multilingual PostASR Text Correction task, reducing the IndicBART model latency from 600ms to 400ms using 4-bit float quantization.
- Fine-tuned the IndicBART model on multilingual data (Hindi, Tamil), comparing PEFT-based LoRA fine-tuning against a regular approach.
- Created model-agnostic inference REST APIs with Locust support for load testing, facilitating internal usage across teams.
Education
B.Tech - Computer Science
IIIT Naya Raipur
Raipur, India07-2022
Senior Secondary -
Sri-Chaitanya
Visakhapatnam, India05-2018
Secondary -
Delhi Public School
Visakhapatnam, India05-2015
Skills
Machine Learning: Representation Learning, Semantic Matching, Named Entity Recognition (NER), Text Classification, Unsupervised Clustering, Agentic AI, Frontier LLM APIs
Tools & Frameworks: FAISS, XGBoost, Airflow, Prism, AWS SageMaker, Google Cloud, Cursor, Claude, Google Antigravity, ONNX, Docker, Locust
Accomplishments
- Quarterly Sprint Velocity Honors: Secured multiple consecutive awards for outstanding performance and high delivery speed during quarters at Interview Kickstart.
- Hult Prize Regional Finalist: Qualified for the international business pitching finale in 2020.
- AP-EAMCET State Rank: Achieved a state rank of 1694 out of nearly 200,000 candidates in 2018.
- Dunnhumby Python Code Combat: Secured a rank of 687 among 5,000+ competitors in the initial round of the 2.0 Hackathon.
Certification
Mathematics for Machine Learning Specialization - Coursera
Work Availability
monday
tuesday
wednesday
thursday
friday
saturday
sunday
morning
afternoon
evening
swipe to browse
Timeline
MLE-II
Interview Kickstart
07.2025 - Current
Data Scientist
Meesho
06.2024 - 07.2025
Associate Data Scientist
Uniphore
07.2022 - 06.2024
Mathematics for Machine Learning Specialization - Coursera
08-2020
Deep Learning Specialization - Coursera
07-2020
B.Tech - Computer Science
IIIT Naya Raipur
Senior Secondary -
Sri-Chaitanya
Secondary -
Delhi Public School