AskMyAstro
An AI astrologer that reads your birth chart and answers real questions over chat. LLM prompt pipeline, chart computation and the full product around it — built and operated solo, end to end.
1.1K+ users · 4.7K views
selected work
Systems in production.
Real users, real traffic, real numbers — my own live products, and platforms built at Stashfin, Zupee and Dresma AI.
personal projects · live
Built solo. Running right now.
Side products I design, ship and operate myself — real traffic, live numbers straight from analytics, and source in the open where it helps.
FileDownloader
A bulk file-download utility serving real traffic on the open internet — paste URLs, get files. Simple tool, sticky usage: over half a million downloads served and counting.
544K+ downloads · 5.9K+ users
Daily traffic — unedited, straight from Google Analytics
Claude Code Switchboard
A local dashboard for every Claude Code session running on your Mac — what each one is working on, live token burn, how much of the plan is left, and the messages flying between sessions. Runs as a plugin or a single command.
MIT licensed · zero dependencies
/plugin marketplace add itssarthak/claudecode-switchboard
/plugin install switchboard
/switchboardcompany work
Platforms built on the job.
No-Code AI Agent Platform
A self-serve platform for building and deploying production AI agents — prompt, tools, model and channels configured from a UI. Cut agent delivery from days to minutes. Today it runs multiple production agents, including a customer-support agent handling ~30,000 messages and 44M tokens per day across 6,000+ daily users.
agent delivery: days → minutes · 44M tokens/day
Centralized LLM Routing Engine
One gateway for every LLM call — budget tracking, rate limiting, multi-provider fallbacks and circuit breaker patterns so product teams survive provider outages without noticing them.
300K+ req/day · 15–20ms · 99% uptime
AI Companion Chatbot
LangGraph multi-agent architecture for response planning and conflict resolution, RAG memory retrieval with Qdrant, and dynamic response timing that simulates human conversation patterns.
46% D15 retention
Dynamic Tool Framework
Turns any REST API into an agent-callable tool: configure the request, fire a live test call to capture the real response shape, annotate the keys — and the LLM tool schema writes itself. Hand-written integrations: eliminated.
any API → agent tool, zero code
RAG Pipeline + Cross-Channel Memory
End-to-end knowledge ingestion — upload, chunk, embed, store — plus a contact system unifying each user's email, phone and Slack identities into a single record with persistent memory across channels.
custom knowledge bases without engineering
Automated Promo Generation
Parallel processing of 50–100 microseries for ad creative, plus a no-code bot management console giving product managers 2-click control over models and behavior.
1 week → 3 hours production time
High-Throughput Processing Architecture
Near-fail-proof architectures for heavy computation workloads on AWS with Kafka-driven microservices — engineered for stability under load.
−40% response time · +50% stability
Running LLMs Locally with Ollama
A practical write-up on self-hosting open-weight models — hardware, quantization and serving trade-offs from hands-on experimentation.