AI Implementation: Team Beyond Limits

Generated from prompt:

#### Slide 1 – Title Slide **Title:** AI Implementation – Team Name :Beyond Limits **Subtitle:** From “Code Monkeys” to “AI Architects” – 24th April 2026 **Bullet points:** Team: Subhra (PM), Papul (QA), Gaurab/Amit/Sagnik(React Native)Rinabrata (React Native & iOS Dev) (React Native), Shreyasi (Python), Suman/Subhamoy/Mayank (Node JS), Soumya/Arindam (React JS) Two solutions: AI‑integrated QA + Dev workflow (using external AI tools) - Three platforms built by us: 1. Multilingual voice tester 2.CodeCanvas **Visual:** Blue/grey gradient + icons --- #### Slide 2 – External AI Tools We Use Daily (Updated) **Title:** External AI Tools We Use Every Day (Our RAG is custom‑built) | AI Tool | Why I chose this | Why NOT alternatives | |---------|------------------|----------------------| | **Codex** | Complex coding logic and refactoring | Gemini: weak multi‑file reasoning | | **ChatGPT** | Best for research, development, and learning new frameworks | Claude: slower iteration for UI code | | **Copilot** | Real‑time inline suggestions inside IDE | CodeWhisperer: less accurate for frontend | | **DeepSeek** | Research, documentation creation, and technical summaries | Perplexity: less depth for architecture | | **Cursor** | Automation script writing (QA) | Copilot: weaker context for test flows | | **Kimi.ai** | Generates PPT reports from execution logs | Gamma: no direct log ingestion | **Antigravity** | Generates APIs & suggests precise, context‑aware code changes | Tabnine: only basic code completion | **Visual instruction:** - Clean table, alternating blue/grey rows, icons per tool --- Here's the **merged and updated slide** combining your original "Before vs After Impact" table with the positive and negative points of using AI in software development. --- #### Slide 3 – Before vs After Impact (Combined QA + Dev) + AI Adoption Pros & Cons **Impact Table** | Task | Before AI | After AI | % Saved | ROI Impact | |------|-----------|----------|---------|-------------| | Code review & refactoring (Codex) | 45 min | 10 min | 78% | Higher quality | | API generation (Antigravity) | 1.5h | 10 min | 89% | Less backend dependency | | Documentation (ChatGPT + DeepSeek) | 1h | 5 min | 92% | Maintainable code | | Automation script writing (Cursor) | 4h | 30 min | 87% | Less manual work | | PPT report generation (Kimi.ai) | 1.5h | 5 min | 94% | Real‑time insights | **Visual suggestion:** Bar charts, green checkmarks for time saved. --- ### ✅ Positive Points of Using AI | # | Benefit | |---|---------| | 1 | Less time, more work | | 2 | Complex code developed in less time | | 3 | Using machine's intelligence | | 4 | Maintain code structure | | 5 | Less human error | | 6 | Faster debugging & automated testing | ### ❌ Negative Points of Using AI | # | Drawback | |---|----------| | 1 | Skill loss over time | | 2 | Developers lack understanding of AI‑generated code despite it being implemented | | 3 | Overconfidence in AI output without proper review | | 4 | Security & compliance risks | --- **Key takeaway for the slide:** AI delivers **significant time savings (78–94%)** but must be adopted with **awareness of skill erosion, knowledge gaps, and security risks**. --- #### Slide 4 – Project 1: Multilingual Voice Tester **Simple workflow (Mermaid):** ```mermaid graph LR A[Teacher uploads PPT] --> B[Student asks in Hindi/Bengali] B --> C[Whisper → English text] C --> D[OpenAI model fixes technical terms] D --> E[**Our RAG** searches PPT] E --> F[Answer generated from PPT] F --> G[Spoken reply in student's language] ``` **Key points:** - No English needed from student - Answer always from today’s PPT Content - Supports Hindi, Bengali, Hinglish, English **Visual:** Blue background, microphone/document icons --- #### Slide 5 – Project 2: CodeCanvas **Simple workflow:** ```mermaid graph LR A[User writes code] --> B{Run} B -->|HTML/React/Vue| C[Instant browser preview] B -->|Python/Java/C++/C / PHP etc. D --> E[Output + errors] C --> F[Live result] E --> F ``` **Key features:** - 20+ languages, no local setup - AI demo mode: natural language → auto‑typed lesson - Used for learning, interviews, prototyping **Visual:** Grey background, code & Docker icons #### Slide 6 – Closing + Call to Action + Thank You **Thank you – Beyond Limits** **Q&A** **Visual:** Big bold numbers, blue “Next steps” box, team photo --- **Final notes for Kimi:** - Mermaid syntax inside markdown fenced blocks - Corporate colours: blue (#1E3A8A) and grey (#6B7280) - Animation: “appear one by one” This fully reflects the external AI tools your team uses daily in development, QA, documentation, and reporting. ""