By Rahul Dhakate · PMP & PSM I Certified · 29 Aug 2026 · learnxyz.in
Most articles about AI for project managers describe what the tools can theoretically do. This article describes what a working project manager and product management leader actually does with them on a typical day — the specific tasks, the specific tools, and the honest reality of how AI fits into a real project management workflow.
Contents
A Real Working Day — How It Actually Flows.
During Development — The Highest-Value AI Use.
Feature Development — Acceleration, Not Replacement
End of Day — Documentation and Communication.
The AI Toolkit — Which Tool for Which Task.
What Genuinely Does Not Work With AI in Daily PM Use.
The Honest Productivity Assessment
My AI toolkit is split approximately 50% ChatGPT and 50% DeepSeek. Claude came into the picture more recently. What I can tell you from direct experience is that the tasks where these tools make the most meaningful difference are development support, content generation, and problem resolution — not as a replacement for PM judgment, but as an accelerator for the analytical and drafting work that surrounds it.
A Real Working Day — How It Actually Flows
A typical day at Artpic.in starts at 10 AM with a meeting focused on goals, delivery checks, and follow-ups across the team. The day moves through coding, testing, and feature work, and ends with an end-of-day meeting reviewing what was accomplished and what carries over.
Here is where AI actually enters that day:
Morning — Before and During the First Meeting
- Review overnight messages and flag any that require complex responses — use ChatGPT to draft responses to the technically complex or diplomatically sensitive ones, then review and send
- If a development issue was raised the previous evening, open DeepSeek with the specific error or challenge before the morning meeting — arrive at the discussion with a proposed solution already researched
- Use ChatGPT to generate a quick meeting agenda if the day’s goals need to be clearly structured for the team
During Development — The Highest-Value AI Use
The development phase of the day is where AI delivers the most concrete, measurable value. At Artpic.in, ChatGPT and DeepSeek came into extensive use specifically for resolving development issues and helping develop particular features quickly.
The pattern: when a developer or I encounter a specific technical problem — an integration issue, a performance problem, an unexpected behaviour in a feature — the first instinct now is to bring the specific error, the relevant code context, and the expected versus actual behaviour to DeepSeek or ChatGPT. The response time from problem to workable solution is dramatically shorter than research-based problem solving alone.
This is not about replacing developer judgment or engineering skill. It is about removing the time spent on research and initial hypothesis generation — the part of problem-solving that used to mean 30 minutes of Stack Overflow, documentation reading, and trial and error. AI compresses that to minutes, leaving the judgment and implementation to the human engineer.
Feature Development — Acceleration, Not Replacement
When we wanted to develop a particular feature quickly, AI enabled us to move faster than the team’s manual capacity would allow. The workflow: describe the feature requirement precisely to ChatGPT or DeepSeek, review the generated approach and code scaffold, adapt it to our specific architecture and standards, then implement.
The time saving on feature scaffolding is significant — initial structure that previously took a developer half a day to produce from scratch now takes 20-30 minutes of AI-assisted generation plus review and adaptation. The quality of the final implementation depends entirely on the developer reviewing and adapting the AI output — AI does not replace that judgment. It removes the blank-page problem.
End of Day — Documentation and Communication
- End-of-day meeting notes: paste raw notes into ChatGPT, ask for cleaned action items and a brief summary formatted for the team channel
- Any stakeholder communication drafted during the day: run through ChatGPT for tone check and clarity improvement
- Risk or issue logs: if a new issue surfaced during the day, use ChatGPT to draft the formal risk entry with description, impact assessment, and proposed response
The AI Toolkit — Which Tool for Which Task
| Tool | Share of Daily Use | Best For | Avoid For |
| ChatGPT (GPT-4o) | ~50% of AI use | Communication drafts, content generation, documentation, meeting notes, general problem-solving | Tasks requiring very recent information without search enabled |
| DeepSeek | ~50% of AI use | Coding problems, technical issue resolution, feature development support, code generation | Long-form business documents — ChatGPT produces more polished output |
| Claude | Emerging use | Complex analysis, long-document work, nuanced writing tasks | Not yet integrated into daily workflow — being evaluated |
| Fathom / Otter.ai | Meeting-specific | Transcription and action item extraction from video calls | Anything other than meeting audio processing |
What Genuinely Does Not Work With AI in Daily PM Use
The most important lesson from daily AI use: knowing what to skip is as valuable as knowing what to use it for.
- Stakeholder relationships — no AI tool replaces the human conversation that builds trust, resolves political tension, or delivers difficult news
- Architecture and strategic decisions — AI can provide input and options, but the decision that carries accountability must be a human decision
- Team culture and motivation — AI can help write a recognition message, but the act of recognising someone authentically is not something you delegate to a model
- Any situation where being wrong carries serious consequences — AI makes confident mistakes. For high-stakes decisions, verify independently
The Honest Productivity Assessment
Since integrating AI tools into daily work at Artpic.in, the most measurable changes are in speed of problem resolution, speed of feature scaffolding, and quality of written communication. The least changed areas are decision-making quality, stakeholder relationship management, and strategic product direction — which is exactly what should be true. AI augments the analytical and drafting work. It does not — and should not — touch the judgment and relationship work.
Start with one task you do every day that currently takes longer than it should. Pick just that one task and integrate AI into it for two weeks. After two weeks you will understand AI’s value in your specific context better than reading any number of articles about it — including this one.
About the Author
Rahul Dhakate is a PMP and PSM I certified project manager and product management leader based in Nagpur, India, with 20 years of experience managing software projects across BFSI, eCommerce, and enterprise software. He uses ChatGPT and DeepSeek daily in his project and product management work at Artpic.in, applying AI primarily to development problem resolution, feature scaffolding, and communication drafting. He writes at LearnXYZ.in about PMP exam prep and AI tools for modern project managers.
Next Article: The Future of Project Management — Predictions for 2026–2030
Good Reads:
