Debasmita Chatterjee

India 0+ yrs exp 67 · Good

About

No bio added yet

Skills & Expertise (32)

Python Intermediate

6.5/10

Years Exp

PyTorch Intermediate

6.0/10

Years Exp

llm apis Intermediate

6.0/10

Years Exp

RAG Intermediate

6.0/10

Years Exp

LangGraph Intermediate

6.0/10

Years Exp

FAISS Intermediate

6.0/10

Years Exp

Hugging Face Transformers Intermediate

6.0/10

Years Exp

REST APIs Intermediate

5.5/10

Years Exp

Git Intermediate

5.5/10

Years Exp

SQL Intermediate

5.5/10

Years Exp

Postgresql Intermediate

5.5/10

Years Exp

Power BI Beginner

5.0/10

Years Exp

GitHub Actions Beginner

5.0/10

Years Exp

Docker Beginner

5.0/10

Years Exp

Apache Spark Beginner

5.0/10

Years Exp

SQLAlchemy Beginner

5.0/10

Years Exp

Prisma Beginner

5.0/10

Years Exp

SQLite Beginner

5.0/10

Years Exp

MongoDb Beginner

5.0/10

Years Exp

WebSockets Beginner

5.0/10

Years Exp

AsyncIO Beginner

5.0/10

Years Exp

Express Beginner

5.0/10

Years Exp

Node.js Beginner

5.0/10

Years Exp

FastAPI Beginner

5.0/10

Years Exp

SDLC Beginner

5.0/10

Years Exp

Debugging Beginner

5.0/10

Years Exp

Testing Beginner

5.0/10

Years Exp

OOP Beginner

5.0/10

Years Exp

Data Structures & Algorithms Beginner

5.0/10

Years Exp

JavaScript Beginner

5.0/10

Years Exp

Typescript Beginner

5.0/10

Years Exp

Scala

Work Experience

AI Safety Evaluator

Mercor

Feb 2026 - Jun 2026

Evaluated AI model outputs for safety, quality, and compliance across diverse AI/ML tasks using structured evaluation criteria. Performed dataset annotation and model evaluation for computer vision and machine learning workflows, identifying errors and improving evaluation consistency.

AI Trainer – LLM Evaluation

Outlier.ai

Aug 2025 - Jan 2026

Evaluated 1600+ LLM responses using structured rubrics to identify hallucinations, reasoning errors, and failure modes, supporting RLHF, model evaluation, and multilingual translation quality assessment workflows. Performed comparative evaluation and preference ranking of multilingual LLM outputs, generating actionable feedback that improved response quality, coherence, alignment, and overall model reliability.