← RETURN HOME
λ FULL SPECTRUM / MISSION INDEX

All work

The full log: every project, case study, and essay — production, research, community, and writing across 28 entries. Filter to narrow the band.

Work entries

Research2026

rocky_mini: A Robot's Brain, Built and Tested Without the Robot

An embodied AI character (Rocky, the Eridian engineer from Andy Weir's Project Hail Mary) built sim-first on a fully local, zero-cost stack. Every external service sits behind a Protocol with a Fake, so the whole four-latency-domain brain runs and passes 126 tests with no robot, no GPU, and no local LLM in the loop. Sim-tested; hardware bring-up pending.

PythonasyncioFastAPIOllamaOPEN LOG →
Writing2026

Will I Hit My Target? A Monte Carlo Planner with ML Inputs

I built a portfolio planner that feeds four machine-learning models into one Monte Carlo wealth simulation, then asks a concrete question. Given a target and a contribution, what are the odds, and how much more would move them? On the synthetic dataset the median lands just short, and the interesting part is what the simulation is made of.

Machine LearningMonte CarloQuantitative FinancePythonOPEN ENTRY →
LOG 004Research2026

No Better Than a Coin Flip: Quantum Kernels on ICU Data

A reproducible quantum-versus-classical benchmark for ICU mortality prediction. Every quantum model lands on the 0.5 chance line while the classical baselines clear it, and the confidence intervals are what make the null result trustworthy.

QiskitQSVMscikit-learnOPEN ENTRY →
Writing2026

Raising an Alien in Simulation: Building a Robot's Brain Without the Robot

I built and tested Rocky's entire software brain in simulation, without the robot, the GPU, or the local LLM in the loop. The honest sim-versus-real boundary is the point, not a caveat, and Protocols with Fakes are what let one codebase run in sim now and drive hardware later.

Embodied AIConcurrencyLocal LLMRoboticsOPEN ENTRY →
LOG 002Research2026

Representation Wins on QA, Not on ML

A paired-data study over 13,800 medication-adherence questions: narrative retrieval wins specialty-med QA, but structured FHIR features win the downstream adherence-prediction task. The representation that wins QA is not the one that wins ML.

PythonFHIR R4BRAGLightGBMOPEN LOG →
LOG 003Research2026

Interventional masked-line ablation reveals shortcut learning

An interventional masked-line audit of a 0.926 macro-F1 LightGBM classifier on Gaia-ESO UVES spectra. F and G read textbook diagnostics; K reads a multi-leg shortcut, and the survey pipeline hides a second one in its continuum.

PythonLightGBMSHAPOPEN LOG →
Community2026

An Anxious Nomad Collective: A Weekly Substack of Book Reviews, Essays, and Long Looks at the Sky

A personal newsletter in the voice the day job does not allow. Book reviews are the entry point. The rest is poetry, personal essay, and slow science writing.

SubstackLong-form EssayBook ReviewsPersonal WritingOPEN LOG →
Research2026

Stellar Spectra with Gradient Origin Networks

The cross-survey data-engineering layer for a Gradient Origin Network project on stellar spectra: survey-specific FITS readers, a ~30,000-star cross-match across APOGEE DR17, GALAH DR3, and Gaia-ESO DR4, and a DVC-tracked HDF5/Parquet store. The generative model itself is upstream; this is the pipeline that feeds it.

PythonPyTorchAstropyGradient Origin NetworksOPEN LOG →
Writing2026

FHIR for AI/ML Engineers: What the Spec Gets Right and Where It Bites

A field guide to FHIR for the ML engineer who just received a half-broken CSV export and a tight deadline. What the standard actually solves, and what it hands you as a research problem in disguise.

HealthcareFHIRML EngineeringData EngineeringOPEN ENTRY →
LOG 007Writing2026

From Physics PDEs to PyTorch

How knowing why a PDE oscillates helps you debug an over-fitting transformer. An essay on transferable intuition between experimental physics and ML.

EssayPhysicsPyTorchOPEN ENTRY →
LOG 001Production2024

Clinical Note Summarizer: MLOps for Healthcare NLP

A FLAN-T5 summarizer fine-tuned on MTS-Dialog, wrapped in a FastAPI service that degrades gracefully, behind CI/CD that authenticates to GCP via Workload Identity Federation.

PythonFastAPIFLAN-T5GKEOPEN LOG →
Research2026

QML-Essentials: 21 weeks of quantum machine learning, honestly graded

A self-paced curriculum that ended with a working quantum autoencoder, four sub-projects of head-to-head comparison against classical baselines, and a clear-eyed view of what quantum ML can and cannot do at simulator scale today.

PythonPennyLaneQiskitPyTorchOPEN LOG →
LOG 006Community2026

Firebird Community Cycle

Board service for a Barrie bike co-op (2022 to 2026): grant writing, POS procurement, and a city-partnered bike-diversion program that keeps hundreds of bikes a year out of the landfill.

Grant WritingOperationsWordPressOPEN LOG →
Community2026

Tech Titans: Teaching Python and FLL robotics at Barrie Public Library

A 24-session community STEM program for ages 9 to 14. Two introductory Python coding cohorts followed by twelve sessions of FIRST LEGO League SUPERPOWERED robotics. Concluded April 2026.

PythonCodedex.ioEduBlocksLEGO SPIKE PrimeOPEN LOG →
LOG 005Community2026

OPLA Psychological Safety Survey

Designed, deployed, analysed and presented a province-wide survey of Ontario public library staff: 1,236 responses across 60+ library systems, measuring 13 psychosocial factors, presented at the OLA Super Conference.

Survey DesignQualtricsStatisticsOPEN LOG →
LOG 008Writing2025

Dismantled by Design

A data-journalism analysis of how coordinated book-banning campaigns are engineered at the policy level, not the parent level.

Data JournalismInvestigativeOPEN ENTRY →
Community2025

Roots of Reality: Volunteer Researcher and Long-Form Writer for a Big History Podcast

Source-vetted research, social-media briefs, and full-length investigative articles for a globally-ranked podcast covering history from the formation of the universe to the present. Volunteer role, since September 2025.

Editorial ResearchSource VettingLong-form WritingSocial Media DraftingOPEN LOG →
Writing2025

Debunking Denialism About Canada's Sixties Scoop Policy

Historical analysis confronting revisionist narratives about the systematic removal of Indigenous children in Canada.

HistoryIndigenous RightsPolicyOPEN ENTRY →
Production2025

Building Healthcare Software at metricHEALTH

FHIR-compliant integrations, automated test frameworks, and clinical KPI dashboards. The work that earned the team Barrie's Mayor's Award for Research and Innovation.

PythonSeleniumFHIRSOC 2OPEN LOG →
Production2025

Summoner: A Plan-Execute-Reflect Agent Framework on GCP

An extensible multi-agent scaffold built on Vertex AI, Cloud Storage, Pub/Sub, and Firestore. The artifact is the orchestration pattern, not the demo task.

PythonGoogle Cloud PlatformVertex AIGeminiOPEN LOG →
Research2024

Quantum vs Classical for ICU Mortality: A Reproducible Benchmark

Three quantum kernel feature maps, a variational quantum classifier, and a quantum neural network, run head-to-head against classical baselines on the WiDS Datathon 2020 ICU dataset. On the reproducible synthetic-fallback run, every quantum model lands on the 0.5 chance line while the classical baselines clear it and train 100x to 10,000x faster.

PythonQiskitqiskit-machine-learningscikit-learnOPEN LOG →
Research2024

Star-Type Classification: A Pipeline I Trust

A six-class stellar classifier in scikit-learn, built around a clean ColumnTransformer pipeline and a reproducible CLI. The artifact is the engineering polish, not the headline accuracy.

Pythonscikit-learnpandasNumPyOPEN LOG →
Production2024

CarePal: AI Wellness Companion for Seniors

Proactive AI companion for senior wellness checks and medication adherence. Built with the Cohere API and a retrieval-augmented LLM in 20 hours. 2nd place at the Georgian College GenAI Hackathon.

Cohere APIRetrieval-Augmented GenerationLLMPythonOPEN LOG →
Research2024

AutoML vs LSTM: Forecasting Iowa Liquor Sales

Comparative study on 19.4 million transactions. Hyperopt-sklearn AutoML reached 85 percent accuracy. An LSTM network reached 88 percent. The interesting finding was the trade-off.

PythonTensorFlowhyperopt-sklearnKerasOPEN LOG →
Research2024

Divvy Bikes: A Year in 4.3 Million Trips

Behavioral and operational analysis of the full 2023 Divvy bike-share dataset, separating member from casual riders and recommending where the operational levers are.

TableauPythonPandasMatplotlibOPEN LOG →
Research2024

A Century of Natural Disasters: Trends, Impact, and Response

Analysis of global disaster records from 1900 to 2021. Droughts and epidemics, less frequent than floods and storms, killed five times as many people.

PythonPandasMatplotlibSeabornOPEN LOG →
Production2024

Real-Time Salary Stream: Scala, Spark Structured Streaming, Kafka, MySQL

Coursework streaming pipeline that ingests employee records over Kafka, classifies them by salary band in Spark Structured Streaming, and sinks the splits to dedicated MySQL tables and downstream Kafka topics. The whole thing runs in Docker Compose.

Scala 2.12Apache Spark Structured Streaming 3.5.0Apache Kafka (Confluent 7.4.0)MySQL 8.0OPEN LOG →
Research2024

Neural Networks from First Principles

Backprop with a pencil before backprop in NumPy. A two-layer network for MNIST-shaped inputs, derived from the chain rule and implemented from scratch. The artifact is the math, not the accuracy.

PythonNumPyJupyterLinear AlgebraOPEN LOG →
© 2026 TIRTHESHJANI.COMOPEN TO AI/ML ROLESAD ASTRA PER ASPERA