cat about.txt

Taranveer Anand_

turning caffeine into side projects

CS senior at Case Western Reserve with a habit of mass producing side projects.

Currently: grinding guitar, and calculating expected value at poker tables.

[view work][contact]
Taranveer Anand
scroll

CURRENTLY BUILDING

$ cat ~/.shipped

shipped_

// mar–may 2026

Apply to jobs while you sleep.

A desktop app that reads a job posting, generates a tailored portfolio project for it, and then applies on your behalf — across workday, greenhouse, lever, and the rest of the ATS zoo.

  • ├──one-click apply across every major ATS
  • ├──auto-generates a tailored project per JD
  • ├──tracks every application end-to-end
  • └──chrome extension for on-the-fly applies
[try shipped →]
job_postagentsubmittedworkdaygreenhouselever

WORK EXPERIENCE

> cat work.log
Heron

Co-Founder, CEO

Jul 2026 - Present

Underwriting for money-back promotions

  • ├──Building an underwriting platform for money-back promotions: a merchant refunds every qualifying customer if a public event happens, and Heron prices that trigger against live prediction markets, hedges the exposure, and keeps the spread.
  • ├──Wrote the campaign calculator that turns capacity, per-head cap, COGS, and margin into a quote, the merchant's projected net, and the promotion's cost expressed as a flat discount.
  • └──Ran go-to-market end to end: 300 sports bar owners contacted, 20 meetings taken, first verbal commitment for season open, and a 949-account qualified lead list across 23 states.
prediction marketsunderwritinghedgingpricinggo-to-market

PulseImaging.AI

Software Engineering Co-Op, part-time

Aug 2026 - Present

Clinical and medical imaging software

  • ├──Building software for AI-driven clinical and medical imaging applications, working with engineers, researchers, and clinical stakeholders.
  • └──Turning functional requirements into shipped features: user interfaces, integrations between application services and databases, plus testing, debugging, and performance work.
healthcaremedical imagingaiinterfacesdatabases
Shipped.one

Founder, Product Engineer

Mar 2026 - May 2026

AI desktop job application platform

  • ├──Built a desktop product that discovers matched roles, tailors resumes, generates portfolio projects, and applies across ATS systems through live browser automation.
  • └──Scaled to 50 concurrent users with application tracking, inbox signal detection, billing controls, and real-time application progress.
electrontypescriptpostgresredisbrowser agents

Aionix.dev

Founder

Jun 2025 - Aug 2025

AI automation systems for U.S. companies

  • ├──Built n8n workflows, custom chatbots, voice AI agents, and CRM integrations to streamline engagement and lead generation.
  • └──Connected webhooks, APIs, CRMs, email, Instagram, and LinkedIn into multi-step outreach and follow-up systems.
n8nwebhooksapiscrmvoice ai

Guru Amardass Private Limited

Database Developer Intern

Sep 2022 - Dec 2022

ERP data migration and reporting

  • ├──Developed Python scripts for data migration and SQL query optimization, reducing processing time by 40%.
  • └──Built data pipelines and dashboards that helped stakeholders debug ERP issues and interpret operational insights.
pythonsqldata pipelinesdashboards

PROJECTS

seednewspolicymarketsim.worldwhat_ifmarket68%forecast

HiveMind

Multi-agent prediction market simulation environment

Simulates agent swarms that research real-world signals, trade against LMSR markets, and surface consensus forecasts through a live dashboard.

Next.jsFastAPILangChainOpenRouterWebSockets
Code
K8s

Flyte

Distributed ML training orchestration for PyTorch on Kubernetes

Wraps torch.distributed, NCCL, and Kubernetes scheduling behind a declarative TrainingSpec with automatic fault recovery and built-in profiling.

PythonPyTorchKubernetesFlytePrometheus
Code
RECOGNIZED

SeeCare

Best Senior Project W25

Privacy-first memory aid for dementia patients

Mobile app using facial recognition and conversation transcription to reduce anxiety and caregiver burden.

FastAPIMongoDBAWSAssemblyAIExpo
sha256sha256sha256rawdedup6 → 3

Cairn

High-performance block deduplication engine in C++17

Rabin-fingerprint content-defined chunking, SHA-256 content addressing, lock-free sharded index. ~500 MB/s chunking, 65% dedup ratio.

C++17OpenSSLCMakeASan/TSan
pricesize105.0020104.0035103.0050102.0025-- spread --101.0030100.004599.004098.0020

OrderBook

Real-time limit order book and matching engine

Price-time priority matching, B-tree backed price levels, Decimal arithmetic throughout. 180-test suite.

Python 3.11Textualsortedcontainersasyncio
Code

Invoke

Universal API-to-tool bridge for Claude

MCP server that lets Claude dynamically discover, research, and register APIs. Ask for data, Claude handles the rest.

TypeScriptNode.jsMCP SDK
Code
[view all projects on github →]

OPEN SOURCE

> gh pr list --author @me --state merged
8pull requests merged

Bug fixes accepted into the main development branches of widely used open-source projects — several merged by project lead maintainers. Each change is small, well-scoped, and ships with regression tests.

[view all merged pull requests on github →]

NOW

cat ~/.now

last updated: aug 2026

├──building heron (useheron.io)
├──grinding NLH-300 sessions
└──learning guitar (badly)

SKILLS

> ls skills/

soft_skills/

how I work

├──product sense
→ Turning vague ideas into usable shipped products
├──technical communication
→ Explaining tradeoffs, APIs, and systems clearly
├──problem decomposition
→ Breaking ambiguous builds into testable pieces
├──rapid prototyping
→ Shipped, Aionix, HiveMind, portfolio experiments
├──ownership
→ Founder-led projects from idea to users
└──cross-functional collaboration
→ Clients, teammates, users, and stakeholders

languages/

what I write

├──python
→ HiveMind, Flyte, SeeCare
├──typescript
→ Shipped, HiveMind, Invoke
├──javascript
→ Frontend and Node.js projects
├──java
→ AI agents, academic projects
├──c/c++
→ Cairn, systems programming
├──rust
→ Learning, systems projects
├──sql
→ PostgreSQL, SQLite, analytics queries
└──bash
→ CLI workflows, Linux automation

ai_search_systems/

agentic and retrieval systems

├──llm orchestration
→ HiveMind, BrainWave, Invoke
├──claude code
→ Agentic development workflow
├──codex
→ CLI-assisted implementation and code review
├──prompt engineering
→ Structured prompts, tool use, and eval loops
├──rag
→ Retrieval-backed assistants and document reasoning
├──embeddings
→ Semantic search and entity matching
├──semantic search
→ Search over docs, profiles, and project data
├──hybrid search
→ Keyword plus vector retrieval patterns
├──multi-agent systems
→ HiveMind simulation and agent markets
├──langchain/langgraph
→ Agent loops and stateful workflows
├──openrouter
→ LLM routing across model providers
├──pytorch
→ Flyte distributed training work
├──scikit-learn
→ ML coursework and data modeling
└──opencv
→ Computer vision pipelines

api_platform_engineering/

connecting systems

├──rest api architecture
→ Product backends and project APIs
├──third-party api integration
→ ATS, CRM, AI, email, and data services
├──api discovery
→ Reading docs and wiring unfamiliar systems quickly
├──webhooks
→ Aionix workflows and event-driven integrations
├──oauth
→ Auth flows and external account connections
├──websockets
→ HiveMind and live product updates
├──background jobs
→ Long-running applies, agents, and automation tasks
├──redis queues
→ Queued work and realtime state
├──browser automation
→ Shipped live form filling across ATS systems
└──cli tooling
→ Developer tools, Invoke, and automation scripts

application_data/

product surfaces and state

├──react
→ Frontend applications and dashboards
├──next.js
→ Portfolio and HiveMind
├──electron
→ Shipped desktop app
├──express
→ Node API services
├──fastify
→ Typed Node backends
├──fastapi
→ HiveMind and SeeCare backends
├──flask
→ BrainWave original backend
├──prisma
→ Typed database access
├──sqlalchemy
→ HiveMind market and agent storage
├──postgresql
→ Relational product data
├──pgvector
→ Vector search in Postgres
├──supabase
→ Auth, storage, and Postgres apps
├──mongodb
→ SeeCare data model
├──redis
→ Queues, cache, and realtime state
└──chromadb
→ Vector database experiments

infra_tooling/

shipping and operating

├──docker
→ Local and deployable services
├──aws
→ S3, Rekognition, and hosted services
├──nginx
→ Reverse proxy and deployment work
├──github actions
→ CI/CD and checks
├──linux
→ Servers and systems coursework
├──n8n
→ Aionix workflow automation
├──vercel
→ Frontend deployments
├──tailwind
→ Fast, consistent UI implementation
└──framer motion
→ Interactive portfolio animations

USES

> neofetch --setup

working setup

editor→VS Code, Claude Code, Codex CLI
terminal→zsh, bash scripts, integrated terminals
os→macOS locally, Linux on servers
workflow→read docs, build small repros, ship the useful part
debugging→logs, network traces, API payload inspection

stack i reach for

├──frontend→Next.js, React, Tailwind, Framer Motion, Electron
├──backend_apis→FastAPI, Express, Fastify, REST, OAuth, webhooks
├──agentic_ai→Claude Code, Codex, LangChain, LangGraph, OpenRouter, n8n
├──data_search→PostgreSQL, Supabase, pgvector, MongoDB, Redis, ChromaDB
├──automation→browser automation, form filling, background jobs, Redis queues
└──infra→AWS, Docker, Nginx, GitHub Actions, Vercel, Linux

CONTACT

> echo $CONTACT_INFO

Want to work together, hire me, or casually argue about something illogical?

>emailtaranveeranand26@gmail.com>githubgithub.com/tarann26>linkedinlinkedin.com/in/tarann26
[download resume.pdf]

// based in san francisco, ca

// ● open to opportunities