Debmalya Sen

Debmalya Sen

+•• ••••• •••••••••••@•••••.com

0.6 yrs

Ask about Debmalya

Skills

Claimed

Matrix CalculationsGenomic Data ProcessingFile Difference AnalysisCommit MonitoringGitLab Actions UsageCode EditingCode SearchVision CommunicationTransparent CommunicationTeamworkVerbal CommunicationWorkflow Design (Process Mapping)

Proof of Experience

Centre for Career Planning and Development (CCPD), NIT Warangal

Training and Placement Associate

Centre for Career Planning and Development (CCPD), NIT Warangal · Apr 2025 – Sep 2025

Publicis Sapient

Engineer Intern

Publicis Sapient · May 2025 – Jul 2025

Publicis Sapient

Software Engineer Intern - Agentic AI Platform

Publicis Sapient · May 2025 – Jul 2025

  • Reduced manual CI/CD configuration effort by 80% and human error by 30%, as measured by internal deployment audits, by designing and delivering a production-grade web application that automated pipeline setup across GitHub and GitLab environments using autonomous LLM-driven tool-use workflows
  • Improved customer support ticket volume by 40%, as measured by weekly support queue metrics, by architecting composable AI agent tools with multi-step task decomposition and autonomous decision-making - collaborating with product managers and security SMEs to align designs with privacy and compliance requirements.
  • Ensured each service layer was independently secure by applying secure design principles and establishing credential-handling invariants across the FastAPI backend, modeling security as code to minimize risk of unauthorized pipeline access.
  • Accelerated team velocity by conducting code reviews for modular backend services, verifying test coverage, diagnosability, and maintainability against agreed engineering standards before production merge.
  • Scaled inference throughput and reduced async task latency by 35%, as measured by load-test benchmarks, by engineering an async Python orchestration layer for AI-driven workflow execution with observable, fault-tolerant request pipelines.

Summer Intern

National Institute of Technology Warangal · May 2024 – Jul 2024

  • Developed an AI-driven model for rapid cancer severity analysis, leveraging thousands of gene networks and metabolic pathways across five major cancer types.
  • Extracted meaningful features from gene expression datasets using deep autoencoders, achieving dimensionality reduction while preserving biological significance.
  • Validated feature clusters using k-means and t-SNE visualizations, ensuring accurate grouping of biologically relevant patterns.
  • Trained a deep neural network on identified features, enabling near-instant cancer severity assessment and reducing computational complexity by 85%.
  • Presented findings at departmental seminars and contributed to drafting a manuscript for submission to a peer-reviewed journal.

Machine Learning Intern - Bioinformatics Research

National Institute of Technology, Warangal · May 2024 – Jul 2024

  • Reduced genomic data processing time by 93.75% across 2M+ gene expression records by re-engineering vectorized statistical modeling modules, cutting overnight batch jobs to under 10 minutes for 5+ PhD researchers; published a production-grade Python package with graph-based gene networking, hypergeometric feature-selection, and automated pathway-scoring pipelines adopted by 5+ external research teams without modification.
  • Collaborated with domain researchers and faculty stakeholders to translate complex biological pathway requirements into precise software specifications, iterating on design options and presenting trade-offs applying modular, independently testable architecture and full documentation to ensure long-term maintainability and reliability.

Proof-of-Work

No completed sessions yet. Anything Debmalya finishes shows up here with the evidence behind it.

Proof of Education

Minor Degree, Computer Science and Engineering (CSE)

National Institute of Technology, Warangal · 2023 — 2026

B. Tech, Biotechnology

National Institute of Technology, Warangal · 2022 — 2026

A Proof CV — one profile, kept current, with the proof attached. Back to the portal