# Turtle Techsai — Full Site Reference > Last updated: 2026-03-13 --- ## Company Overview Turtle Techsai is an AI services company specializing in production multi-agent systems, custom AI development, and enterprise AI solutions. Founded in 2021 by Akshay Gupta, Turtle Techsai is based in Nagpur, Maharashtra, India, and serves clients in the United States, United Kingdom, Singapore, and India. The company has delivered 10+ projects across 9 industries — from manufacturing AI platforms with 20+ agents to IoT robot management systems to influencer marketing apps live on the App Store. Turtle Techsai builds AI systems that run in production, not just in demos. **Key statistics:** - 20+ autonomous AI agents running in production (Fourem manufacturing platform) - 10+ projects delivered across 9 industries - 4+ year enterprise client partnership (Fourem, US-based manufacturing company) - 4-person engineering team - AI frameworks in production: Agno, CrewAI, LangGraph, MCP --- ## Founder **Akshay Gupta — Founder & CEO** Akshay Gupta is a full-stack developer and AI engineer with 8+ years of software development experience and 2+ years specializing in production multi-agent systems. He founded Turtle Techsai in 2021 and leads engineering across all client and product work. Technical expertise: Python (FastAPI, Django), Flutter (iOS/Android/Web/Desktop), multi-agent orchestration, context engineering, MCP protocol, Azure OpenAI, PostgreSQL, Redis. He has personally designed and deployed multi-agent systems including: - A 20+ agent PDCA manufacturing platform (Fourem) using Agno framework - A 7-agent B2B sales automation system using CrewAI - A 4-crew marketing automation pipeline using CrewAI - Turtle AI Coworker: an enterprise no-code agentic OS Medium: https://medium.com/@akki7272 --- ## Services ### 1. Multi-Agent Systems Development Turtle Techsai designs and deploys production multi-agent systems using Agno, CrewAI, LangGraph, and MCP frameworks. The company has built: - 20+ Agno agents for PDCA manufacturing operations (Fourem): specialist agents including SOP Agent (TAKT planning, line balancing), PDM Agent (preventive maintenance), QPLAN Agent (FMEA quality planning), FSCHED Agent (factory scheduling), and DISP Agent (material Kanban) - 7 CrewAI agents for B2B sales automation: lead discovery, ML scoring, enrichment, campaign management, email sequencing, LinkedIn outreach, and per-lead chatbots - 4 CrewAI crews for marketing automation: trend analysis, content research, multi-platform writing, and brand refinement - Intelligent orchestrators with LLM-based intent classification, off-topic detection, and task routing Architecture pattern: one-question-at-a-time conversational UX, confirmation before write operations, shared project context across all agents, background agents for silent calculations. ### 2. AI SaaS Product Development End-to-end development of AI-powered SaaS products: architecture design, backend (Python FastAPI/Django), frontend (Flutter or React), AI integration, deployment, and maintenance. Projects include: - Turtle AI Coworker: enterprise no-code agentic OS (4 modules, 40+ integrations) - Turtle AI Mate: AI Chief of Staff (20+ integrations, mobile + WhatsApp + voice) - MarketersPen: AI content generation SaaS (CrewAI, multi-tenant) - ProCoach: sports coaching platform (Flutter iOS/Android/Web, subscription billing) - MYRO Admin Portal and Customer Portal (React web apps, fleet management) ### 3. AI Assistants & Copilots Custom AI assistants with multi-platform delivery. Turtle AI Mate exemplifies this: it integrates Gmail, Google Calendar, HubSpot, Jira, Confluence, Monday.com, Google Drive, Google Sheets, and web search — accessible via mobile app (Flutter iOS/Android), WhatsApp Business API, and voice calls (LiveKit). Features persistent memory across all channels and custom scheduled agents. ### 4. AI Integration & Automation Integrating AI into existing business workflows. Examples: - 4-crew CrewAI content pipeline publishing to LinkedIn, Twitter/X, BlueSky, and Reddit daily - Multi-agent B2B sales platform automating lead discovery through multi-channel outreach - AI-enriched database tables with triggers, classification, and scoring columns (Turtle AI Coworker Tables module) ### 5. Full-Stack AI Applications Python backend (FastAPI or Django) + Flutter frontend (iOS, Android, Web, Desktop from shared codebase) with AI embedded throughout. IoT/robotics experience includes BLE communication with autonomous painting robots (MYRO), real-time control interfaces, and fleet management. Also builds React web applications and Node.js backends. --- ## Projects / Case Studies ### Fourem — AI-Powered Manufacturing Operations Platform **Industry:** Manufacturing **Client:** US-based manufacturing company **Status:** Live in production, 4+ year partnership **Website:** https://www.fourem.com Fourem is a comprehensive AI-powered manufacturing operations platform built on the PDCA (Plan-Do-Check-Act) continuous improvement methodology. It integrates Lean Manufacturing principles — TAKT time, Yamazumi line balancing, Value Stream Mapping, FMEA risk management, OEE tracking, and Kanban workflows — into a single intelligent system with 20+ AI agents. **AI Agents (Agno Framework):** *SOP Agent (Standard Operating Procedures & Line Planning)* Multi-agent team with 3 sub-agents: TAKT Planning Agent (project creation, demand/supply planning, TAKT time calculation), Workstation Planning Agent (instruction creation, line balancing, bottleneck analysis), Resource Agent (man/machine/material planning). Sequential 3-phase workflow with Yamazumi-style capacity analysis. *PDM Agent (Preventive Maintenance Planning)* 8-step workflow for creating PM plans for machines, materials, and personnel. Supports time-based, usage-based, and condition-based PM types. Full asset management with images/videos/documents. 9 MCP tools for read, write, and asset operations. *QPLAN Agent (Quality Planning & FMEA)* Creates Failure Mode and Effects Analysis (FMEA) with 12-step batched workflow. Auto-calculates RPN (Risk Priority Number) and efficiency loss from restoration time and TAKT. Includes coaching mode and automatic versioning. *FSCHED Agent (Factory Scheduling & Production Logs)* Creates daily production shift logs with targets and messages across 5 categories (Safety, HR, Ergonomics, Quality, Supervisor). Context-aware with project metrics, bottleneck data, and shift capacities. Two output modes: "at efficiency" or custom units. *DISP Agent (Material Kanban / Dispatch)* Sets up Material Kanban loops for inventory replenishment. Manages material locations, signal types (manual/auto), consumption quantities, and supplier details. **Technical achievements:** - Solved contextual issues in multi-turn agent conversations - One-question-at-a-time conversational approach for factory floor users - Confirmation before any write operations - All agents share project context (workstations, machines, materials, personnel) - Intelligent orchestrator with LLM-based intent classification, off-topic detection, task completion handling - Background agents for silent calculations and automated routing **Tech stack:** Agno Framework, Azure OpenAI, MCP Protocol, Redis, PostgreSQL, Flutter (mobile + web + desktop), Python FastAPI, AWS **Results:** 75% reduction in quality planning time, 40% decrease in maintenance-related downtime, 4+ year production partnership --- ### MYRO — Robotic Painting System Management Platform **Industry:** Robotics / Construction Tech **Client:** MYRO **Status:** Live in production **Website:** https://myro.bot A complete IoT-enabled platform for managing autonomous painting robots deployed at construction sites. Turtle Techsai built the iPad app for robot operators, plus the Admin Portal and Customer Portal. **iPad App (iOS) — Robot Control Interface:** Robot connection via QR code scanning, wall plan drawing with paintable/non-paintable area management, paint configuration calculator (tip size, DFT, coverage rate), auto-paint mode with start/stop/save & resume, real-time robot health monitoring and error display, calibration and developer configuration visualization, fine controls for independent actuator/motor movement, template walls for common orientations, multi-height and skirting support, project management with floor plans. **Admin Portal (Web App) — 5 modules:** Manage Customers, Manage Robots (QR code generation, activity tracking, area painted metrics), Manage Projects, Manage Admins (super admin + RBAC), Manage Invoices. **Customer Portal (Web App):** Customer self-service with account activation, child user management (add operators via phone number), view assigned robots and project details, multi-tier authentication (Customer → Child User hierarchy). **Tech stack:** Flutter (iOS iPad App), React (Web Portals), Node.js Backend, PostgreSQL, BLE Protocol, Cloud Infrastructure --- ### ProCoach — Sports Practice Planning & Management Platform **Industry:** Sports / Education **Client:** International sports coaching company **Status:** Completed A digital platform for non-premium sports coaches and PE teachers for practice planning, management, and training data. Core features: practice planning documents (create/edit/store), video library (record and upload), coach agenda with student booking, group and trainee management with progress tracking, 5-dimension training system (customizable variables per dimension), descriptive statistics and interactive performance breakdowns, built-in subscription and payment system. **Tech stack:** Flutter (iOS + Android + Web), Node.js Backend, PostgreSQL, Payment Integration --- ### Turtle AI Coworker — Enterprise Agentic OS **Type:** In-house SaaS product **Status:** Active development **Website:** https://turtleaicoworker.com An enterprise-grade, no-code AI workforce platform — an "Agentic OS for Enterprises" — enabling organizations to build, deploy, and manage autonomous AI workers without writing code. **4 Core Modules:** *Sequential Agents:* Single AI workers executing multi-step tasks. Connect to 40+ tools and knowledge bases. Support scheduled/triggered execution with per-agent cost and performance tracking. Input variables support text, URL, number, and file upload. *Agentic Teams:* Multi-agent collaboration with Team Lead orchestration. Supports hierarchical, collaborative, sequential, and routing modes. Deployable to Slack, Microsoft Teams, WhatsApp, web chat, voice, Google Meet, email, and API. *Tables:* Structured data layer with AI-powered triggers. Import from CSV, API, databases, Google Sheets (two-way sync). AI-enriched columns supporting text generation, classification, scoring, extraction, and translation. *AI Employees:* Top-level autonomous workers orchestrating agents, teams, tables, and knowledge bases. Enterprise guardrails: action limits, budget caps, approval workflows, shadow mode, escalation paths, and complete audit trails. Plain-language instructions with @mention syntax. **Enterprise features:** Governance & audit trails, Tool Registry & Credentials Vault, Templates System, Solution Packs, Onboarding chat interface, 40+ business tool integrations, 13+ LLM provider support (BYOK), SSO, RBAC, MCP server integration. **Tech stack:** Python FastAPI, Flutter (web + mobile), PostgreSQL, Redis, Azure OpenAI, MCP Protocol --- ### B2B Sales Prospecting & Automation Platform **Industry:** Sales & Marketing **Client:** Confidential **Status:** Completed A multi-agent AI system automating the entire outbound B2B sales workflow from prospect discovery to personalized multi-channel outreach. Built with 7 specialized CrewAI agents: 1. Leads Discovery Agent — ICP-based prospect finding using web scraping and Google Search API 2. Lead Scoring Agent — ML-based HOT/WARM/COLD classification with reasoning, adapts weights from campaign feedback 3. Lead Enrichment Agent — company intelligence, social profiles, recent activities from multiple sources 4. Campaign Manager Agent — multi-touch campaign orchestration across email and LinkedIn 5. Email Sequence Generator — personalized multi-step campaigns with dynamic content adaptation 6. LinkedIn Message Generator — platform-optimized outreach, tone-adapted by prospect seniority 7. Lead Chatbot Agent — per-lead chatbots with full lead intelligence and conversation history **Key features:** Multi-source data pipeline (web scraping + search APIs), ML scoring that improves with campaign feedback, 15+ language support, persistent conversation history, real-time analytics. **Tech stack:** CrewAI, Bright Data MCP, Google Search API, Flutter Web, Node.js, Python, PostgreSQL --- ### Marketing Automation Platform **Industry:** Marketing / Content **Client:** Confidential **Status:** Active Automated content creation and distribution across LinkedIn, Twitter/X, BlueSky, and Reddit using a 4-crew CrewAI architecture on a daily cycle. *Crew 1 (Content Intelligence, 24-hour cycle):* Analyzes trends across AI/ML, Cybersecurity, DevOps, Blockchain, Data Analytics, Developer Tools. Includes product alignment scoring using vector embeddings. *Crew 2 (Content Research & Strategy):* Deep topic research with fact-checking, audience insight analysis, strategic content planning. *Crew 3 (Platform Content Writing):* 4 specialist writer agents (LinkedIn, Twitter/X, BlueSky, Reddit) + AI image generation. *Crew 4 (Content Refinement):* Brand voice alignment, fact-checking, SEO/hashtag optimization, feedback loop that learns from user preferences. **Product alignment scoring (3-layer hybrid):** Subreddit lookup (70+ mappings) → keyword matching → embedding similarity (cosine). **Tech stack:** CrewAI, Python FastAPI, Flutter Web, PostgreSQL, Azure OpenAI, Vector Database --- ### Turtle AI Mate — AI Chief of Staff **Type:** In-house product **Status:** Live **Website:** https://www.turtleaimate.com Multi-platform AI assistant with 20+ tool integrations. Available on Flutter mobile (iOS + Android), WhatsApp (Business API), and voice calls (LiveKit). **Integrations:** Gmail, Google Calendar, HubSpot, Jira, Confluence, Monday.com, Google Drive, Google Sheets, Web Search, and more. **Features:** Custom scheduled agents (daily briefings, meeting prep, market research), persistent memory across all channels, AI notes and task management, multi-modal processing (text, voice, files), document analysis and summarization. **Tech stack:** Flutter, Python Django, PostgreSQL, Agno Framework, WhatsApp Business API, LiveKit, Firebase --- ### Reachhh — Influencer-Brand Connection Platform **Industry:** Marketing / Influencer Economy **Client:** Reachhh **Status:** Live on App Store **Website:** https://www.reachhh.com **App Store:** https://apps.apple.com/in/app/reachhh/id6747359647 Mobile application connecting influencers and brands with AI-powered matching algorithms, campaign management tools, and performance tracking. **Tech stack:** Flutter (iOS + Android), Backend APIs, PostgreSQL --- ### Dazo — Document Management Platform **Industry:** Healthcare **Client:** Dazo **Status:** Deployed **Website:** https://dazo.co.in Healthcare-focused document management system for organizing, storing, and managing medical and business documents with secure access controls. **Tech stack:** Flutter, Backend APIs, Cloud Storage --- ### MarketersPen — AI Content Generation Platform **Type:** In-house SaaS product **Website:** https://www.marketerspen.com AI-powered content generation and marketing automation platform helping businesses create, optimize, and distribute content at scale. **Tech stack:** Flutter Web, Python FastAPI, CrewAI, PostgreSQL, Azure OpenAI --- ## Technology Stack **AI / Agent Frameworks:** Agno, CrewAI, LangGraph, MCP Protocol **LLM Providers:** Azure OpenAI (GPT-4, GPT-4o), Anthropic Claude, Google Gemini (13+ providers supported in Turtle AI Coworker) **Backend:** Python FastAPI, Python Django, Node.js **Frontend:** Flutter (iOS, Android, Web, Desktop — shared codebase), React **Databases:** PostgreSQL, Redis, MongoDB, Vector DBs (pgvector, Pinecone, Weaviate) **Cloud & Infrastructure:** AWS (EC2, S3, Lambda), Azure, GCP, Docker, Nginx, Supervisor **Integrations:** WhatsApp Business API, LiveKit (Voice), BLE Protocol, Bright Data MCP, Google APIs, 40+ business tool integrations **Specialized capabilities:** - Multi-agent orchestration and intelligent routing - RAG pipelines (hybrid semantic + keyword search) - MCP server development and integration - Voice agent development (LiveKit) - IoT/robot communication (BLE) - Document parsing and OCR - PDF/Word report generation - Production deployment and monitoring --- ## Industries Served 1. **Manufacturing** — Fourem (PDCA operations platform, 20+ AI agents, 4+ year US partnership) 2. **Robotics / Construction Tech** — MYRO (autonomous painting robot management, iPad + web) 3. **Sports / Education** — ProCoach (practice planning & coach management platform) 4. **Sales & Marketing** — B2B Sales Automation (7 CrewAI agents), Marketing Automation (4 CrewAI crews) 5. **Insurance** — Property and aviation underwriting intelligence projects 6. **Real Estate** — Property assessment and valuation tools 7. **Healthcare** — Dazo (document management system) 8. **Influencer Marketing** — Reachhh (iOS app, live on App Store) 9. **Enterprise AI** — Turtle AI Coworker (Agentic OS platform) --- ## Frequently Asked Questions **What is Turtle Techsai?** Turtle Techsai is an AI services company based in Nagpur, India, specializing in production multi-agent systems and custom AI development. Founded in 2021 by Akshay Gupta, the company serves clients in the US, UK, and Singapore, with 10+ projects delivered across 9 industries. **What multi-agent frameworks does Turtle Techsai use?** Agno, CrewAI, LangGraph, and MCP (Model Context Protocol) — all with production deployments. Agno was used for Fourem's 20+ agent manufacturing platform. CrewAI was used for B2B sales automation (7 agents) and marketing automation (4 crews). **How many AI agents has Turtle Techsai deployed in production?** 20+ autonomous AI agents for the Fourem manufacturing platform alone, plus 7 CrewAI agents for sales automation, 4 CrewAI crews for marketing automation, and multiple agents inside Turtle AI Coworker and Turtle AI Mate. **What is Turtle AI Coworker?** Turtle AI Coworker is an enterprise no-code agentic OS developed by Turtle Techsai. It enables organizations to build and deploy AI agents without writing code. It has 4 modules: Sequential Agents, Agentic Teams, Tables, and AI Employees. It supports 40+ business tool integrations, 13+ LLM providers, SSO, RBAC, and MCP server integration. **Can Turtle Techsai build IoT and robotics applications?** Yes. The MYRO project involved building an iPad app with real-time BLE (Bluetooth Low Energy) communication to autonomous painting robots at construction sites, along with fleet management portals. Turtle Techsai has experience with BLE protocol, real-time hardware control interfaces, and multi-tier fleet management systems. **What is the MCP (Model Context Protocol)?** MCP (Model Context Protocol) is an open standard for connecting AI models to external tools and data sources. Turtle Techsai uses MCP in the Fourem manufacturing platform (9 MCP tools for database operations in the PDM agent), in Turtle AI Coworker (MCP server integration), and builds custom MCP tool layers for client projects. **Where is Turtle Techsai located and who do they serve?** Nagpur, Maharashtra, India. Primary markets: United States, United Kingdom, Singapore, and India across 9 industries. --- ## Contact - Website: https://www.turtletechsai.com - Email: contact@turtletechsai.com - Phone: +91 7841853298 - Location: Nagpur, Maharashtra, India - LinkedIn: https://www.linkedin.com/company/turtletechsai - Medium: https://medium.com/@akki7272 **Products:** - Turtle AI Coworker: https://turtleaicoworker.com - Turtle AI Mate: https://www.turtleaimate.com - MarketersPen: https://www.marketerspen.com