Adapting a Literary Classic for Screen, Pitch Document by AI — Turning *The Age of Goodbyes* into a Platform Adaptation Proposal with Claude Code

The Malaysian film A Love Letter to Grandma is about to open. There’s a great deal of delicate emotional writing in it, and it’s genuinely moving.

It made me think of a landmark work of Malaysian Chinese literature — Li Zishu’s novel The Age of Goodbyes. Same subject matter, same emotional delicacy. And it got me thinking today: if The Age of Goodbyes were adapted for the screen, how would I build its pitch document from a producer’s seat?

So this episode, I’m building that pitch deck end to end with Claude Code, and walking you through everything I did today.

Step 1: have it run deep research

I opened Claude Code and had it stand up a team of agents to do deep research across the web.

Roughly what I asked for: get a full picture of the novel’s content and its reception online; analyze which angles could make it a hit if adapted for screen; analyze its audience and which streaming platform suits it; do a casting analysis too — which actors fit which roles; analyze what pacing and strategy the marketing should use; and analyze the benchmark titles in the same genre from recent years. Do a first wide-open sweep across all of that and present it to me as a web page, in the PPVI visual style.

If I had the full text of the novel on hand, all the better — I could have it work from the whole book. I don’t, so I only had it find the synopsis and coverage online. Doing this for real, you’d give it the entire novel.

What the research page looked like

Once it finished collecting, the page I got back was themed around whether The Age of Goodbyes could be adapted into a hit, examined across twenty-odd adaptation angles.

It found the Douban rating and the reputation that came with being one of Asiaweek’s top ten novels of the year; it analyzed which elements could carry a hit adaptation — its emotional anchor, the multilingual texture of the Nanyang setting, the layered ensemble of women — all of which help it break out and generate conversation. It also pulled benchmark titles from literary adaptations to compare against, like CCTV’s A Lifelong Journey, Tencent’s Blossoms Shanghai, and To the Wonder, explaining why benchmarking against those is the steadier, more proven path.

On platform fit, my own first thought was iQiyi or Tencent Video too — a premium mid-length series of around twenty episodes with serious literary underpinnings is exactly right for those platforms. Add the audience analysis, the things to watch out for in adaptation, and which types of clips to use for marketing across various channels, and it had a first pass at all of it. On casting, for the male lead it suggested Greg Hsu, Bai Yu, and Zhu Yawen; for the female roles it suggested Wen Qi — all reasonable. Audience: women roughly 20 to 40.

Distilling my pitch-document experience into a skill

Because I used to make these pitch documents constantly, I built a skill: I gave it the relatively mature pitch decks I’d made before and had it learn the architecture, the length of each section, and which information sources you pull the data from to assemble a pitch.

Then my prompt was roughly: I’ve given you all the research documents on adapting The Age of Goodbyes for screen; now I need a roughly 40-page deck to take this literary work’s screen adaptation to a platform; use my pitch-document skill and fill out everything from the one-line hook, the synopsis, the project highlights, and the reputational backing of the existing IP, through to character bios and casting; use GPT image 2 to make it an all-image deck; the whole project tells a story of Nanyang overseas Chinese, so use a period film-stock filter throughout; generate the images in parallel for a first version; and since the casting is settled, generate look-test photos of real performers for the top choice in each role.

That’s all it took — one simple prompt — and it mobilized all my accumulated project experience to produce the deck and finally convert it into a PDF pitch document.

I made two rounds of edits along the way: first, some pages didn’t give me a detailed synopsis, so I asked it to write two dedicated pages; and the first pass couldn’t produce good look-test photos because of performers’ likeness rights, so I gave it about two more rounds of feedback. From data collection to final generation, because it drew on my earlier pitch-document experience, the whole process went very smoothly.

The finished pitch document

Open the PDF it made, and there’s the pitch document.

There’s a key-art cover setting the overall tone; the one-line positioning is “She cannot see, yet she sees a whole city clearly” — because the work tells the story of an entire city’s ensemble of characters through the eyes of a blind woman. Project overview: a 20-episode web series, roughly 45 minutes an episode, a female coming-of-age period realist drama, set in Malaysia, spanning 1970 to 2018.

It broke out the source material’s value and the IP’s reputational backing: one of the top ten Chinese-language novels of 2020, winner of the Malaysian Chinese Literature Award, Douban rating of 9.1. Below that, the author’s bio, the case for greenlighting it (why choose a serious literary work like this for adaptation now), the historical setting, the synopsis (an ensemble story of ordinary Chinese Malaysians living in a cheap public-housing block in the tin-mining city of Ipoh), the protagonist’s core story arc, the core draws, the character relationships, and the main character bios — all built out page by page. The thematic statement: “the one who cannot see is the one who sees through the world and everyone in it.”

After that come the strategy for expanding the ensemble in adaptation, the values, its differentiated position in the current Chinese drama landscape, corroboration from benchmark titles (A Lifelong Journey, To the Wonder, Blossoms Shanghai, The Long Season, all high-reception premium dramas), and the target audience (a quality drama, 25 to 45 skewing female, with spillover into overseas distribution to Southeast Asian Chinese communities). For key creatives, the director it ultimately recommended was the Malaysian filmmaker Lau Kek-huat — though the director doesn’t have to be Malaysian, plenty of directors in China could do this well. Plus a screenwriter recommendation, lead casting (Wen Qi, Zhou Dongyu, Zhang Zifeng), and the veteran character actors for the ensemble (Wu Yue, Kara Wai, Angelica Lee). It closes with commercial value and marketing strategy.

The result is already striking, and I don’t think it falls short of the quality of the distribution pitches I used to make.

How much further this could go

There are of course details worth adjusting. I’d probably have AI research further and produce a longer list of which other performers fit this project; once a budget is set, you can also go find key creatives against the budget.

One more important point: casting decisions now also consider whether a performer has a history of scandal, and Nanyang subject matter will inevitably involve Hong Kong and Taiwan actors, so you can use AI to run due diligence on performers — check whether they’ve had bad press before, and use that as a screen.

Overall, as long as we distill and extract the experience and thinking of a veteran producer or executive producer, AI now clears the bar on a pitch document — hand it a novel and out comes a complete screen-adaptation plan, fast.

This genuinely is AI changing life, AI changing the shape of an entire business.

See you next episode.

Source: EP0028_audio.mp3 · ASR model gemini-2.5-pro (chunked parallel) · full text of the roughly 11.5-minute original recording

[00:00] You really can take a novel and then produce a complete screen-adaptation concept and plan for it, very fast. A Love Letter to Grandma (给阿嫲的情书) has taken off, and that film has so much finely observed emotional expression in it, it’s genuinely moving. And it’s about to open in Malaysia. Which got me thinking about a landmark work of Malaysian Chinese literature, a novel called The Age of Goodbyes (流俗地) — same kind of subject matter, also very finely observed emotional expression.

[00:25] So today I couldn’t help wondering: if The Age of Goodbyes were adapted into a screen production, then from a producer’s point of view, how would I put together the pitch document? So today I’m going to take you through building that pitch document with Claude Code. Let me recreate the entire process I went through today making this pitch-document PPT. First I open Claude Code and have it do deep research. There’s a work of Malaysian Chinese literature,

[00:50] a novel called The Age of Goodbyes. And I think this work has enormous potential to be adapted into a hit screen production — especially as a premium mid-length series of around 20 episodes. So I need you to stand up a team of agents and do deep research across the whole web. First, get a comprehensive understanding of the novel’s content and its overall reputation and reviews online. And then research and analyze for me: if we adapt it into a screen production, which entry points —

[01:16] would help turn it into a hit. And at the same time I need you to analyze its audience and which streaming platforms it suits for release. Including the main characters — you can help me with casting, meaning which actors would suit playing these roles. That needs analysis. And at the same time analyze what kind of rhythm and strategy the marketing rollout should use.

[01:41] Including its — do an analysis first of comparable works, benchmark works in similar territory from the last few years. So first we collect all of this in a divergent sweep, and then present the results of that collection to me as a web page, using the PPVI visual style. So it did a divergent pass first. If I had the full text of the novel, that would be even better, of course — you could have it read the whole synopsis of the novel —

[02:06] and have it write that out first. But I don’t have it right now, so I only had it go online and find some synopses or introductions to the novel itself, and understand the work that way. When we actually do this for real, we’d hand it the whole literary manuscript of the novel. And then I also had it comprehensively research the adaptation potential, because the existing readership of the literary work matters a lot too, and its reputation matters a lot. And then I also had it do competitive — what you’d call research on works of the same type. So once that —

[02:34] content was collected, what I got at the time was a web page like this: can The Age of Goodbyes be adapted into a hit? And like I said, that’s looking at it as an adaptation of about 20-odd episodes. And in here it includes — it found the Douban rating, and looked at its reputation, that it was one of Asia Weekly’s top ten novels of the year. And then, in adapting it into a hit, which points can actually support it becoming a hit — including its emotional anchor, including its —

[03:00] multilingual Nanyang ecosystem, including its audience targeting, that stratified ensemble of women. All of these points help it break out, help it generate talkability, help it build its hit poten— boost its hit potential. And then including — it also pulled out some benchmark works of pure-literature adaptation, like CCTV’s A Lifelong Journey, and like Tencent’s Blossoms Shanghai, and To the Wonder. It benchmarks against these pure-literature adaptations, and why works like those —

[03:26] represent a safer, proven benchmark path to success. Including platform fit in here. Honestly, the first one I thought of was probably iQIYI or Tencent Video too, especially for a premium series of maybe 20 episodes. It’s especially suitable — especially a work with this kind of serious-literature foundation is better suited to airing on a platform like iQIYI or Tencent Video. And including the audience analysis, and the things to watch out for in —

[03:51] the adaptation. It pulled all of that out too, including which types of clips you need when running the marketing across multiple channels, or what approach to take for that rollout. It gives you a preliminary idea on all of it. And including the casting in here — the core roles, the male lead. Look, it recommended Hsu Kuang-han, Bai Yu, Zhu Yawen, which are all pretty solid, actually. And then for the female roles, look, it recommended people like Wen Qi. These are all pretty suitable —

[04:16] as actors. And then including for a 20-year-old audience — actually the whole female demographic from about 20 to 40 is the audience. And its marketing playbook is all in here. So once I take this material and process it and hand it to Claude Code again — this is a demonstration — the next step is that I have it produce a —

[04:41] what’s called a pitch document. Because I used to make these pitch documents all the time, I actually built a skill for it, where I hand it the relatively mature pitch-document PPTs I’ve made before, and it learns the architecture in them, learns the section-by-section lengths, and learns which information sources it can pull the data from to assemble a pitch document like that. So I’d build a skill and have it produce the PDF. As for the rough prompt, here’s how I put it today:

[05:07] I’ve already given you all of our research documents on the screen adaptation of The Age of Goodbyes. Now we need to make a pitch-document PPT of roughly 40 pages, to pitch platforms on a screen-adaptation plan for this literary work. And you can use my pitch-document skill, which fully covers everything from the one-line logline, the story synopsis, the project highlights, then the reputation backing of the existing IP,

[05:33] including the character bios across the board, and then the actor casting. So let’s produce a PPT like that. And you can use GPT image 2 to make an image-only PPT. As for the project’s overall style, you know it’s a Nanyang project telling an overseas-Chinese story, so I think overall we can go with a period-feeling film-stock filter style —

[05:59] for the overall visual direction. And I need you to generate the images in parallel. Let’s put out that first version. At the same time, since I think our casting is settled too, I need you to take the first-choice ideal casting and generate the character look-test stills using the real performers. So a simple prompt like that, and it can draw on my existing project experience —

[06:24] to produce this PDF — PPT, ultimately converted to PDF — pitch document. Now, in today’s build I actually went through two rounds of revisions, because when it did it the first time, some pages’ content — the first time through it didn’t give me a detailed story synopsis. So I gave it feedback saying I needed it to write out two dedicated pages of story synopsis. And then including, on the first generation, because of performers’ portrait rights,

[06:49] it couldn’t generate good look-test stills. So I actually did about two more rounds of feedback. So today, from data collection to generating the final pitch document, because it drew on all my earlier pitch-document experience, it took roughly an hour-plus to finish building this pitch document. And let me show you the final result. You can see this PDF is the pitch document it made. This is the overall tonal —

[07:14] key-visual cover, the one-line positioning: she couldn’t see, yet she saw a whole city clearly. Because the work itself tells the story of a whole city’s ensemble of characters through the eyes of a blind girl. The project overview: it’s a 20-episode web series, roughly 45 minutes an episode, a female coming-of-age, period-leaning realist drama. It’s set in Malaysia, running from 1970 to 2018, so it’s a multi-decade drama.

[07:41] And then the value of the source novel and the IP backing — look, it specifically lists that it was one of the top ten Chinese-language novels of 2020, and then that it won the major Malaysian Chinese literature prize, and its Douban rating is 9.1. So there really is very strong reputation behind the source. And then an introduction to the author, including the case for greenlighting it — why now is the time to pick a serious-literature work like this for a screen adaptation. And then its period setting, its story synopsis, which is: in a cheap block of public-housing flats in the tin city,

[08:07] this ensemble story of all kinds of ordinary Chinese Malaysian lives unfolding. And then including the description of the protagonist’s core story line. And then why — what its core appeal is, an analysis of why audiences would want to follow this series. Including the character relationships in it, and the bios of the main characters, all laid out page by page. And then, like, the thematic intent in here is about how the person who can’t see is the one who sees straight through the ways of the world. And there’s an enormous amount of fine-grained —

[08:34] human emotion in there, including how the emotion is expressed. And then its strategy for expanding the ensemble in adaptation — there’s analysis on that here too. Its values, including the values within the adaptation strategy, and then how it uses characters to thread the story’s through-line. And then its — its subject matter’s distinctive, differentiated positioning within Chinese drama as it stands. And then its benchmark works — the selling points mentioned earlier actually get corroborated in those benchmark works too.

[09:00] Including A Lifelong Journey, including To the Wonder, including Blossoms Shanghai, The Long Season — these are all high-reputation premium dramas you can benchmark against on certain traits. As for its target audience, as mentioned, it’s a quality drama, so 25 to 45, skewing female. And because it’s a Nanyang-themed piece, it can actually spill over into overseas distribution markets. And the ethnic Chinese market in Southeast Asia is also a very suitable audience segment. And for the key creatives, the director — look at who it ultimately recommended,

[09:25] Lau Kek-huat, who’s a Malaysian director. But you don’t have to pick a Malaysian director. I’m sure there are plenty of directors in China who could do a great job on a work like this. And including the recommendations for these key creative team members, and then the screenwriter, the recommendation for the adapting screenwriter. The leads you want — Wen Qi was mentioned earlier, and people like Zhou Dongyu, Zhou Dongyu, Zhang Zifeng, right? Those are all pretty suitable candidates, actually. And then the male roles you want,

[09:50] including the ensemble roles picked in here — I still think they’re pretty good. People like Wu Yue, and then Kara Hui, and Angelica Lee, all veteran heavyweights who’d be perfect for this film. And then its commercial value, including some simple marketing strategy, all of it you can see in this pitch document. Of course, this is the result of a pitch document made in just over an hour, and it’s already stunning. I think this is already no worse in quality than the distribution pitch documents I —

[10:15] used to make back in the day. Of course there are also some details to adjust in here. And I might go further and use AI to research more, have it look for and list a longer roster of which other matching actors would suit this project. And once the project’s budget is set, you can also go back and look for the key creative team based on that budget. Actually one thing I think matters a lot: these days when a screen production casts, they also consider whether a performer —

[10:40] has a history of misconduct. Especially with a Nanyang drama like this, you’re bound to have some Hong Kong and Taiwan performers in it. So you can also use AI to run due diligence on a batch of performers, right? Whether they’ve ever had bad press about related things — you can use that as a screen. So overall, I think the level AI can now hit on pitch documents — as long as we take the experience and thinking of a seasoned producer or executive producer —

[11:06] and distill it, extract it out, then we really can take a novel and then produce a complete screen-adaptation concept and plan for it, very fast. This really is AI — AI changing life, AI changing the whole shape of the business. That’s it for today. See you next episode, bye-bye.