Home A - Z
Page Index

Blame

764228 Pooja Thorat 2025-10-12 09:32:51 1 
# Enovate IT-QMS-PL01-Project Plan-[FlowCast]-V1.0
2
3
# **Tools & Services**
4
5
1. Gemini Flash 2.5/Gemini Flash 2.5 Lite/Gemini Flash 2.5 Pro
6
1. Code analysis
7
2. Steps and flow generation
8
3. Video analysis
9
4. Generation of timestamps, subtitles, transcripts
10
5. Generate FFMPEG script to edit the video
11
2. Gemini embedding model
12
1. Steps and flow embedding
13
3. Gemini Flash 2.5 TTS OR Tortoise TTS
14
1. Voiceover generation
15
4. Langchain to build automatic agents that will handle standardized communication with the AI models
16
5. Minio docker to store video files, audio files, uploaded code
17
6. Postgres database
18
7. Postgres plugin pgvector for vector database OR Qdrant vector database OR Weaviate vector database
19
8. **Paid service required [Gemini Developer API Paid Tier](https://ai.google.dev/gemini-api/docs/pricing)**
20
21
# **Project Flow**
22
23
1. Setup Flow
24
1. Cypress/Playwright repo uploaded by QA/project owner
25
2. AI indexes the code-base and generates the steps being performed by the automation
26
3. AI creates vector embedding for the steps and uses it as the source of truth
27
4. Generate Vector embedding and metadata for the scripts
28
1. Store the vector embedding into the vector database
29
2. Store the metadata about all the steps into the database
30
3. Store the code into storage bucket
31
2. Prompting Flow
32
1. User/QA Input Prompts(ex: Generate demo for adding user to org ) / Prompts through API
33
2. Check if the video for the requested flow exists using filenames and metadata
34
1. (YES)Serve the video
35
3. Check if the requested flow exists in the script/repo
36
1. (NO)Inform that the video for such flow cannot be generated in a positive manner
37
4. Runs the automation script in the server
38
1. Use video output flags to generate video for the automation
39
5. Video output along with metadata is saved
40
6. AI Agents
41
1. Analysis for the video by AI
42
2. Generate data for events:
43
1. Timestamps
44
2. Locations
45
3. Type
46
3. Generate transcripts, subtitles and voice-over for the video
47
4. Generate FFMPEG command/script to combine the video overlays, voice-over, subtitles to generate new video
48
7. Generate the edited demo video
49
8. Output the demo video and serve to the user
50
3. [URL](https://excalidraw.com/#json=4i66sXvJ1zdvc_WPLEtsK,okBD43RRjL2DifWRSahj6g)
51
be7119 Melisha Dsouza 2025-11-11 07:39:57 52
![](./image-1762846732917.png)
764228 Pooja Thorat 2025-10-12 09:32:51 53
be7119 Melisha Dsouza 2025-11-11 07:39:57 54
![](./image-1762846744010.png)
764228 Pooja Thorat 2025-10-12 09:32:51 55
56
# **AI Parts of the Challenge**
57
58
1. Code understanding and generating vector embedding and metadata for the flows
59
2. Video analysis (events, transitions, etc)
60
3. Video metadata generation
61
4. Transcripts, subtitle, voiceover generation
62
5. FFMPEG script/command generation
63
64
# **Team plans and goals**
65
66
1. First 2 weeks goals to achieve:
67
1. Finalise the backend architecture (monolith/microservices, vector db, cache, docs, ai tools)
68
2. Finish the boilerplate of the backend
69
3. Remove unnecessary elements from the UI and finalize the UI
70
4. Have some testing scenarios for the qa ready
71
2. First month goals to achieve:
72
1. Have all the public and private apis ready
73
2. Have the code upload \-\> indexing \-\> embedding flow ready (at least in some capacity)
74
3. Have all the test scenarios ready and begin testing for the code upload flow
75
4. Start working on the video generation and editing part
76
77
# **Sequence Diagram**
78
be7119 Melisha Dsouza 2025-11-11 07:39:57 79
![](./image-1762846759609.png)
764228 Pooja Thorat 2025-10-12 09:32:51 80
81
# **ERD**
82
be7119 Melisha Dsouza 2025-11-11 07:39:57 83
![](./image-1762846768656.png)
764228 Pooja Thorat 2025-10-12 09:32:51 84
85
# **Technology Stack**
86
87
1. Frontend: React
88
2. Backend: Microservice of Java and Python
89
3. LLM: Gemini 2.5-FLash, gemini-embedding-001
90
4. AI Orchestration: LangChain
91
5. Integrations: FFMPEG, Automations
92
6. DB: Postgres (with pgvector)
93
7. Deployment: Docker, Gitlab CI
94
95
#