Skip to content
View trishxaaa's full-sized avatar
:dependabot:
:dependabot:

Highlights

  • Pro

Block or report trishxaaa

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
trishxaaa/README.md

Trisha Mukherjee

I build things because I want to understand why they work, not because a tutorial told me to.

Most of my work lives somewhere between research, engineering and data science. I'll read a paper until I find the unanswered question underneath it, then build the smallest experiment that can prove (or disprove) the idea. Sometimes the hypothesis survives. Sometimes it doesn't. Both outcomes are useful.

I'm naturally drawn to systems. Give me a single model and I'm already thinking about retrieval pipelines, evaluation frameworks, orchestration, observability, and the trade-offs that appear once something leaves a notebook and has to behave consistently.

My curiosity doesn't really stop at AI. One week it's representation learning or medical vision models. The next it's the psychology of why humans exist, designing a better Notion system for my life, or disappearing down a rabbit hole that absolutely wasn't part of the original plan.

I like repositories that explain themselves. Code matters, but so do the decisions behind it. If you've scrolled past the installation instructions and you're reading the README because you want to know why something was built a certain way, you're exactly the audience I had in mind.

Currently Exploring

  • Grounded multimodal AI
  • Evaluation for LLM systems
  • Medical vision-language models
  • Agentic workflows
  • Whatever paper I read at 2 AM

Still chasing interesting questions.

Selected Publications

• IEEE ASIANCON 2025 — Boosting LLM Performance on Boolean QA Using Cluster-Based Embeddings

• IEEE FMLDS 2025 — Evaluating the Impact of Input Format on LLM Performance

• IEEE iSSSC 2025 — ECG-based Biometric Encryption for Remote Patient Monitoring

Find Me Elsewhere

Always happy to talk about research, engineering, or an interesting rabbit hole!

LinkedInGoogle ScholarEmail

Pinned Loading

  1. career-compass career-compass Public

    AI-powered multi-agent career guidance platform leveraging LangGraph, FastAPI, React, Supabase, and LLMs for personalized career roadmaps, resume analysis, and workflow automation.

    TypeScript

  2. mentra-prp-ai-agent mentra-prp-ai-agent Public

    AI-powered PRP assistant built with FastAPI, LangGraph, React, Supabase, and Groq LLM for personalized career guidance and workflow automation.

    TypeScript

  3. vlm-radiology-report-generation vlm-radiology-report-generation Public

    Vision-language model for automated chest X-ray radiology report generation using clinically grounded visual and textual representations.

    Jupyter Notebook

  4. nlp-robustness-benchmark nlp-robustness-benchmark Public

    Comparative robustness evaluation of NLP architectures under linguistic perturbations on the BoolQ question-answering benchmark.

    Jupyter Notebook

  5. three-parameter-ged-simulation three-parameter-ged-simulation Public

    Statistical estimation and Monte Carlo simulation of the Three-Parameter Generalized Exponential Distribution using MPS, MMLE, and LSPF.

    Jupyter Notebook