AI Agent Engineer Projects in New DelhiAI Agent Engineer Projects in New DelhiPromptOT – AI Prompts Get Refined, Versioned, Evaluated & Shipped
PromptOT is a prompt management platform designed to help AI teams treat production prompts as production code. It lets teams author prompts in structured, typed blocks, version every change with full history and rollback, evaluate versions against saved test cases across multiple models, and deliver the compiled, variable-driven prompt to their application via a single API call or native MCP integration, with no redeploy required.
We built a compilation engine solid enough for production use, an AI co-pilot for conversational prompt editing with inline diffs and scoring, and native support for the tools AI teams already use daily - Claude Desktop, Cursor, ChatGPT, Codex CLI, Windsurf, and Zed.
Key Features -
Typed Prompt Blocks
Semantic Versioning with Rollback
Evaluations Across Models
API & MCP Delivery
AI Co-Pilot for Prompt Editing
AI teams often struggle with -
Prompts scattered across a Google Doc, a Slack thread, someone's Notion, and hard-coded strings in the codebase
No version history, no diffs, no way to know which version is actually live
No way to evaluate a prompt rewrite before shipping it to production
Legal and brand review happening informally in DMs, if at all
PromptOT delivers a single source of truth for every production prompt, shipped by API or MCP. It bridges the gap between prompt experimentation and reliable, production-grade delivery, turning prompts from fragile prose into managed, versioned infrastructure.
Where most Sales Manager go wrong isn't lack of data, it's getting overwhelmed with numbers and targets. What actually matters is not numbers, but "why" behind it.
So we built an Agentic AI Sales Engine that sits on your Telegram, dissects the "why" behind what's happening in your sales team. It doesn't just throws number but tells you why someone is falling short while another pulling ahead, surfaces the operational bottlenecks underneath the numbers, and suggests what can actually be done better, all in real time at 0 infrastructure cost.
Problem:
A Sales Manager needed a fast way to check teams performance, calls, leads, conversions, pipeline value, but without opening spreadsheets or chasing manual reports. But raw numbers alone don't tell you what to do. Two reps can have identical conversion rates for completely different reasons. The manager needed something that could reason about the data, not just report it but by answering questions like "who needs attention today?" or "why is Rahul underperforming?"
What We Built
A fully automated Telegram bot, powered entirely by a self-hosted n8n workflow, that reads live data from Google Sheets and combines two layers: hard KPI reporting on demand, and an AI reasoning layer that interprets those numbers into a story a manager can act on.
The "Why Layer"
This is a part that makes it more than a dashboard, instead of just telling Rahul 12% conversion, the engine reasons over calls, leads pipeline and conversion patterns together and drives an analysis thereby providing the right suggestive next steps.
Skills Demonstrated
:Workflow automation & API orchestration (n8n)
:OAuth 2.0 debugging and Google Cloud API setup
:Data processing / aggregation logic (JavaScript in Code nodes)
:LLM integration with grounded, hallucination-resistant prompting
:Conversational bot design (Telegram Bot API)
:Building production-usable tools on a strict zero-cost budget Now, meet the one I already told you all my project jarvis which is complete now. Here's the details -
JARVIS is a production-grade, privacy-first AI desktop assistant designed to operate fully locally on standard hardware. Built to rival modern desktop AI systems, JARVIS seamlessly integrates natural voice interaction, real-time screen vision comprehension, deep Windows OS automation, and intelligent document/presentation generation.
Unlike basic wrappers around cloud APIs, JARVIS features a hybrid architecture combining zero-latency regex intent routing, local LLM fallbacks, Win32 API shell controls, and active VLM screen verification.
Key Capabilities of Jarvis -
🎙️ Multimodal Voice & Audio Intelligence
Bilingual STT & Dynamic Query Cleaning: Real-time speech recognition tuned for Hinglish, Hindi, and English with automatic phonetic filler word stripping.
Expressive Local TTS & Emotion Effects: Low-latency neural speech synthesis powered by Piper ONNX and Edge TTS with adaptive prosody and emotional modulation.
Hands-Free Media & Non-API Automation: Full hardware media key automation for Spotify and browser video playback without requiring paid API tokens.
👁️ Vision AI & Live Screen Comprehension
VLM Window & Screen Verification: Captures active window frames using OpenCV and local vision models (Moondream / Qwen-VL/Mistral) to verify OS tasks (e.g., verifying opened folders, app states, or UI elements).
Camera Emergency Sentinel: Real-time visual distress sentinel using multimodal vision checks before initiating priority emergency calls.
🖥️ Deep OS & File System Automation
Subfolder Inspection & Bulk Purging: Inspects complex nested folder structures (e.g., Pictures/Screenshots), calculates storage footprints, and executes secure file/folder purges via Win32 shell calls.
OneDrive-Aware Name-Based Resolution: Intelligent 3-tier lookup engine resolving standard paths (Desktop, Downloads, Pictures) across native paths and OneDrive redirects without needing absolute user paths.
Silent Recycle Bin Clean & Disk Optimization: Win32 API integration (SHEmptyRecycleBinW) for 100% silent, error-free disk maintenance.
📄 Productivity & AI Document Generation
Automated Presentation Engine: Generates styled PowerPoint presentations (.pptx) with custom slide themes, topic summaries, and automated asset downloads.
Markdown & PDF Document Compiler: Built-in Marp compilation engine converting voice notes to polished PDF slides and documents.
And not only that I have open-sourced the entire github repo you can install it, check it and run it to your laptop as your assistant too, and don't forget to give the star, and if you face any issue kindly dm me or message in github too.
Here's the link - "https://github.com/darshitp091/Jarvis " 
