Skip to content
pyd.log

Prasana

Generalist engineer who figures it out and delivers.Fintech · Quant · Full-Stack.

Download ResumeGet in Touch
Scroll
Entry 001

About

I'm Prasana — a generalist engineer who doesn't believe in being one-dimensional. I build machine learning models, automate data infrastructure, and ship full-stack applications. When something needs to get done, I figure it out and deliver — whether I've done it before or not.

Right now I'm interning at HSB (a Munich Re company) where I've built prediction models that cut error rates by more than half and automated pipelines saving 150+ person-hours monthly. I study Computer Science at the University of Waterloo and I'm targeting roles in fintech, quant, and big tech.

Beyond software, I have a deep interest in equities and investment vehicles, and I'm a NAUI Master Scuba Diver with 120+ logged dives and Teaching Assistant credentials. I believe good software starts from trying to solve a real problem — not from picking a shiny tech stack.

Entry 002

Deep Dives

Model dashboard screenshot

Insurance Limit Prediction Model

Context

At HSB (Munich Re), underwriters relied on manual heuristics to set insurance limits — a slow, inconsistent process that left money on the table and introduced risk.

Approach

Built a LightGBM ensemble model with SHAP explainability, fed by a cleaned pipeline of 50k+ policy records. Designed feature engineering around loss history, exposure metrics, and industry codes. Iterated weekly with underwriting stakeholders to calibrate outputs.

Outcome

Reduced prediction error from 19% MAPE to 7% MAPE — adopted by the underwriting team as the default recommendation engine.

MAPE improvement19% → 7%
Records processed50,000+
PythonLightGBMSHAPSQLAzure ML
Pipeline architecture diagram

Automated Reporting Pipeline

Context

Monthly reporting at HSB required analysts to manually pull data from multiple sources, format spreadsheets, and update Power BI dashboards — consuming 150+ person-hours monthly.

Approach

Designed an end-to-end Python pipeline integrating SQL Server, Azure Data Factory, and Power BI APIs. Built parameterized report templates and a scheduling layer that runs automatically on the first of each month.

Outcome

Saved 150+ person-hours per month and eliminated manual formatting errors. Reports now auto-generate and land in stakeholder inboxes by 9 AM.

Hours saved monthly150+
Error rate→ 0 manual errors
PythonSQL ServerAzurePower BIPandas
Entry 003

Index

Stock Analysis DashboardMulti-security analysis across 44+ equities with ML predictions
PythonPlotlyscikit-learn
↗
Reddit Sentiment PipelineReal-time sentiment analysis using DistilBERT on subreddit data
PythonHuggingFacePostgreSQL
→
Data Anonymization ToolPII detection and masking engine for insurance datasets
PythonPresidioAzure
→
Entry 004

Transmissions

Career

A Generalist's Guide to Technical Interviews

When your background spans ML, full-stack, and quant — how do you prepare for interviews that want you to be a specialist? My approach after 30+ mock interviews.

View all posts→
Entry 005

Beyond the Terminal

When I'm not building software, I'm probably 30 meters underwater or reading 10-Ks. I hold NAUI Master Scuba Diver and Teaching Assistant credentials — which means I've spent enough time managing risk at depth to know that preparation isn't optional. I have a strong interest in equities and have built analysis tools tracking 44+ securities with ML-driven predictions. I think the best engineers are the ones with range — people who bring perspectives from outside the codebase.

Entry 006

Coordinates

Elsewhere