ShubhamTiwari

Real-time derivatives platform with sub-250ms ML-driven execution.

• Engineered a real-time derivatives trading platform with an async market-data pipeline aggregating tick-level data into OHLC candles via WebSocket streaming with <250ms latency and Protobuf serialization.
• Designed 6 specialized ML models with online calibration and drift detection; built a risk engine enforcing per-trade (1% equity) and daily drawdown (3%) limits, validated across 1,200+ backtesting sessions.

Distributed job-automation engine with GPT-4 resume tailoring.

• Built a distributed job-automation system scraping 1,500+ listings/day from LinkedIn, Indeed, and Naukri with deduplication, normalized pipelines, and 4 specialized Celery queues backed by a Redis broker and dead-letter handling.
• Implemented GPT-4 powered resume tailoring per job description with LaTeX-based version control, match scoring, and a Playwright auto-fill engine with human-like delays and CAPTCHA detection.

AI that QA-tests web apps like a senior engineer, then files the tickets.

• Built a 7-package monorepo where AI agents autonomously crawl web apps, detect JS errors, broken flows and accessibility issues (axe-core), then auto-generate bug reports and Playwright regression tests.
• Multi-tenant SaaS with Jira/GitHub ticket filing, Stripe billing tiers, and swappable LLM backends (OpenAI, Anthropic, Ollama).

Multi-agent autonomous trading system for the Indian markets (NSE).

• Designed a 4-agent pipeline — Scout (screening), Market Analyst (live candles), Risk Manager (9-point validation) and Trade Executor — coordinated over a Redis event bus.
• Powered decisions with local LLMs (Ollama qwen3) + GPT-4o-mini; a live Next.js dashboard streams trades and candle updates over WebSockets.

Local-first Windows assistant with a modular multi-agent brain.

• Wake-word assistant that transcribes fully offline via Vosk and routes intents to 12+ specialized agents (apps, web search, media, files, system commands).
• Config-driven agent registration with safety confirmations gating destructive actions like delete, move and shutdown.

Chat with your files — 16 cooperating agents managing every drive.

• Natural-language file management across all drives — organize, dedupe, analyze storage and clean up — orchestrated by a 16-agent pipeline (Scanner, Classifier, Duplicate Detector…).
• Concurrent multi-threaded scanning with a TTL-based SQLite cache and safe Recycle-Bin operations; Ollama LLM with keyword fallback.

GPU-accelerated video frame-interpolation desktop app (RIFE v4.25).

• Multiplies video FPS 2×/4×/8× and downscales 4K→2K using RIFE optical-flow interpolation on a 3-stage concurrent Decode → GPU → Encode pipeline.
• Preserves all audio, subtitle and chapter tracks with NVIDIA NVENC hardware encoding (HEVC/H.264/AV1) for RTX 20/30/40 GPUs.

Autopilot that creates and publishes content across 6+ networks.

• Generates short-form videos, images, captions and hashtags, then schedules and publishes to Instagram, TikTok, Facebook, X, LinkedIn and Google Business.
• Multi-tenant with Stripe subscriptions and a BullMQ worker queue orchestrating parallel content generation and publishing.