Workflow Hub

I’m a very unsophisticated user—and an older man—so my use of ChatGPT is still fairly basic. But it has changed my life. I use ChatGPT as a research assistant for a book I’m writing. In practice, what I do is not much more than an in-depth search, but ChatGPT has taken me places in this journey that I would have thought impossible just three months ago. We’re now deeply engaged in solid research that will likely last about two years before we reach the writing phase. I’m energized by what we’re doing and wanted to share the experience. I believe AI is going to change the way working historians conduct research. Step-by-step: 1. I use ChatGPT as a research assistant for the book I’m writing. 2. I use it for in-depth searching and research. 3. I continue developing the research with ChatGPT over an expected two-year period before beginning the writing phase. 4. I reflect on how this process is changing my research journey and the way historians may conduct research.

Build a Tiny HTML Research Logger Instead of Fighting a Spreadsheet Turn repeated research measurements into a simple phone-friendly interface instead of making participants wrestle with spreadsheets. Some research data is simple. Collecting it often isn't. Imagine recording pain, energy, mood and a few additional variables several times during a study. A spreadsheet can technically do this. But on a phone? Open the file, find the right row, locate the columns, enter values without accidentally changing something, save and close. That's a ridiculous amount of interface for a tiny measurement. If recording how you feel takes ten seconds mentally but two minutes administratively, the tool has become part of the research burden. So I asked: What is the smallest interface someone actually needs when logging an observation? My answer was a tiny HTML page that works on both phone and desktop. 1. Separate the interface from the dataset Participants shouldn't need to interact with rows, columns or database structures. They should see a few simple fields: Pain, Energy, Mood, Optional Note and Save. Behind the interface, each submission becomes a structured record containing the measurements, a timestamp and any optional context. The researcher gets organised data. The participant gets simple controls. 2. Define the research measures first Before asking AI to build anything, establish what you're measuring. Which variables matter? What does each scale mean? Which values are allowed? Should measurements remain comparable over time? These are research decisions, not interface-design decisions. AI can help implement your measurement system, but shouldn't quietly invent the methodology. 3. Design for repeated use A questionnaire completed once can tolerate some friction. A logger used several times a day cannot. Make controls large enough to tap comfortably. Use readable labels, vertical layouts and obvious actions. Avoid unnecessary navigation, tiny cells and repeated setup. Design for the phone first, then let the interface adapt to larger screens. And test the fifth entry, not just the first. Something that looks elegant during development can become irritating when you're tired, distracted or recording yet another observation. The ideal repeated action is almost boring. That's a compliment. 4. Automate the invisible work Participants shouldn't have to enter information the system already knows. A logger can automatically generate timestamps, record identifiers and technical metadata. Each submission should become one coherent record. Show a clear confirmation when saving succeeds and prevent accidental duplicates. Keep optional notes available without turning every measurement into a diary assignment. Don't make participants do the database's job. 5. Keep the interface neutral A research logger is a measurement instrument. Its colours, labels, ordering and controls shouldn't subtly suggest which answers are desirable. Likewise, the logger doesn't need to interpret someone's measurements or announce that their mood seems lower on Tuesdays. That's analysis, not collection. First gather observations reliably. Interpret them later using appropriate research methods. 6. Decide where the data lives A single HTML file can provide the interface without requiring a full application. But HTML alone doesn't provide persistent storage or synchronisation. Browser storage may work for a local prototype, although it has limitations. For research involving several devices, participants or longer periods, you'll need suitable storage and export options. A phone and laptop displaying the same interface don't automatically share observations. Keep the dataset consistent and exportable, for example as CSV. For sensitive information, include appropriate privacy and security safeguards. 7. Build only what's necessary It's tempting to expand a tiny logger into an entire platform. Dashboards! Charts! Profiles! Notifications! Accounts! Perhaps later. Start with Open → Measure → Save. Test the process repeatedly on phone and desktop. Improve whatever causes genuine friction. You can enlarge buttons or rearrange the layout without changing the study. But changing a measurement scale is a research-design decision, not a cosmetic update. The workflow Define the variables and scales → ask AI to build a responsive HTML logger → add automatic timestamps and structured records → choose suitable storage and export methods → test repeated logging → refine usability without changing the measurements. Why this works A spreadsheet is excellent at displaying a dataset. That doesn't make it the best place to create one. Researchers think in variables, records and analysis. Participants simply need to answer the question in front of them. Those shouldn't have to be the same interface. Sometimes the best research tool isn't a sophisticated platform. It's a tiny page that gets the spreadsheet out of the participant's way.

I built three interconnected websites for the Solomon for Jesus ministry. The first, solomonforjesus.com, houses the shining jewel: an AI chatbot named Solomon. The second, solomonforjesus.net, is a ministry academy with four levels of instruction and 16 lessons per level. The third, solomonforjesus.net, is home to my book *Fingerprint of Reality* and other articles and essays. All three websites are still in development, but they are operating and publicly available. I use GitHub repositories for coding, Vercel for hosting, Cloudflare for domains, and Supabase for back-end security. Step-by-step: 1. I built three interconnected websites for the Solomon for Jesus ministry. 2. I used solomonforjesus.com for the Solomon AI chatbot. 3. I used solomonforjesus.net for the ministry academy, which contains four instructional levels with 16 lessons per level. 4. I used solomonforjesus.net for my book *Fingerprint of Reality* and other articles and essays. 5. I use GitHub repositories to code, Vercel to host, Cloudflare for domains, and Supabase for back-end security. 6. I keep all three websites operating and publicly available while they remain in development.

I regularly use ChatGPT for lesson planning and materials development with students ranging from EFL learners to UK undergraduates. I keep a separate anonymous thread for each student, so GPT can track the context and suggest the best direction for the next session. I also contribute my experience as a tutor while using AI to create fresh worksheets based on typical materials from various UK exam boards. Today, I had a successful interview for a lecturer role. Gemini helped me design and plan my demo seminar, and it looks like I’ve landed the job. Step-by-step: 1. I create a separate anonymous thread for each student. 2. I use ChatGPT to track each student’s context and suggest the best direction for the next session. 3. I combine AI-generated ideas with my experience as a tutor when planning lessons and developing materials. 4. I use typical materials from various UK exam boards as the basis for creating fresh worksheets. 5. For my lecturer interview, I used Gemini to design and plan my demo seminar. 6. I had a successful interview, and it looks like I’ve landed the job.

I built a suite of teaching tools that work independently and together. Chalkline supports maths, while Penline supports literacy. Both tools provide quick access to the New Zealand curriculum for planning and feed information directly into my planning utility, Planline, making teacher planning faster. Planline integrates with Google Drive and is set up with Stripe billing for subscriptions. The suite also includes Compass World News, a news aggregator with a freemium model and a pro unlock. The apps are designed without user tracking. Findaroo is a freemium word-find game with ads. Slashkit helps Minecraft players quickly create commands to use in games. Everything is built into my website, www.danielsoutham.co.nz, which is hosted on Cloudflare. I push updates and changes through GitHub, so I can make changes—big or small—from anywhere, on any device. Changes are then pushed and merged in one prompt. Step-by-step: 1. I built Chalkline for maths and Penline for literacy, with each tool able to work independently. 2. I connected both tools to the New Zealand curriculum so teachers can quickly access it for planning. 3. I connected Chalkline and Penline to Planline so they can feed information into the planning utility. 4. I integrated Planline with Google Drive and set up Stripe billing for subscriptions. 5. I added Compass World News as a freemium news aggregator with a pro unlock and no user tracking. 6. I added Findaroo as a freemium word-find game supported by ads. 7. I built Slashkit to help Minecraft players quickly create commands for use in games. 8. I brought all of the apps together on www.danielsoutham.co.nz and hosted the website on Cloudflare. 9. I use GitHub to push and merge updates, allowing me to make changes from anywhere, on any device, in one prompt.

I submit two reports to my directors each month. Both use structured templates, but the updates come from multiple people and sources. Some information is unstructured, such as hallway conversations that are not always captured in email, so I have struggled to get the accuracy I need when using AI to build the reports. I am currently testing CoWork and Copilot Work IQ. These reports used to take me hours, so I am already saving time, but I want to refine the process further. I believe a better way to save and store the source data could improve the final outputs, although I have not worked out the right approach yet. Step-by-step: 1. I collect updates from multiple people and sources for the two monthly reports. 2. I organize the information into the structured templates required by my directors. 3. I use CoWork and Copilot Work IQ to help build the reports. 4. I account for unstructured updates, including information from hallway conversations that may not have been captured in email. 5. I am exploring better ways to save and store the source data so I can improve the accuracy of the reports while continuing to save time.

I built a workflow to examine complaints I had about a local bureaucratic service and check whether my account of what happened was supported by the written record. I was concerned about delays, incomplete handling, and the way my case progressed after I gave feedback about a harmful aspect of the service and asked for a different person to handle the dossier. That request was refused. Afterwards, I felt that the process had become slower, more confusing, and more prone to mistakes. Rather than asking GPT to confirm that the service had handled things badly, I used it to test my own interpretation against the evidence. I asked GPT to review the relevant emails and dossier communications, reconstruct the chronology, and identify requests, replies, delays, unresolved issues, and changes in how the case was handled over time. We then compared the periods before and after my feedback and request for reassignment. The central questions were deliberately neutral: Were my complaints grounded in the correspondence? Did the handling of the dossier objectively become slower or more incomplete afterwards? Were there concrete errors or unanswered questions in the record? Or was I remembering the experience as worse than the documentation supported? This distinction mattered to me. The workflow was not designed to produce a verdict or turn frustration into evidence after the fact. It was designed to challenge my assumptions and separate what I felt from what could actually be demonstrated. Where the documentation supported a complaint, we could point to the relevant chronology, delays, unanswered questions, or inconsistencies. Where the evidence was incomplete or ambiguous, that was noted too. The result was a more grounded account of the case: not “I know this was handled terribly,” but “these are the parts of my experience that are supported by the written record, these are the parts that remain uncertain, and this is where the timeline changed.” In simple terms, the workflow was: emails and dossier communications → chronology reconstruction → before-and-after comparison → review of delays, errors, and unresolved issues → evidence check against my complaints → a grounded account of what can and cannot be supported. What I liked about this workflow was that it used AI as a reality-checking tool rather than an agreement machine. It helped me test whether my criticism was actually anchored in the record before I relied on it in further communication. Step-by-step: 1. I gathered the relevant emails and dossier communications about my case. 2. I asked GPT to reconstruct the chronology and identify requests, replies, delays, unresolved issues, and changes in how the case was handled. 3. I compared the period before and after my feedback about a harmful aspect of the service and my request for reassignment, which was refused. 4. I examined whether the record showed that the process became slower or more incomplete, and whether it contained concrete errors or unanswered questions. 5. I separated points supported by the documentation from points that remained incomplete or ambiguous. 6. I used the resulting chronology and evidence check to create a grounded account of what my complaints could and could not demonstrate.

A while ago, my dad told me he wanted to use AI to analyze World Cup matches. He had never coded before, and his AI experience was mostly limited to occasionally asking Gemini a question. I installed Claude Code for him and gave him a few prompting tips. He ended up building a full football analytics site himself and deploying it on Vercel so he could show it to his friends. Getting started with AI is easier than people assume. Sometimes, you just need a project you genuinely want to make. Now that the Premier League season has kicked off, he’s already reworking the site for it. Step-by-step: 1. I helped my dad choose a football analytics project he genuinely wanted to build for analyzing World Cup matches. 2. I installed Claude Code for him, since he had never coded before. 3. I gave him a few prompting tips and let him use AI to help build the site. 4. He built a full football analytics site himself. 5. He deployed the site on Vercel so he could show it to his friends. 6. After the Premier League season kicked off, he began reworking the site for the new competition.

AI can make a child’s drawing look more polished, but it can also erase exactly what made it interesting. A yellow tree, a rabbit larger than a person, a horse with impossible legs, or a tiny floating object may look like mistakes to a model—but they may be the whole point. I use the child’s drawing as art direction: preserve its visual logic, then build outward. The child supplies the invention; AI helps develop the surroundings. Step-by-step: 1. I read the drawing before transforming it by making a neutral description of what is visible: placement, relative sizes, colours, expressions, unusual shapes, and small details. For example: “A large yellow tree stands on the left. Two tiny figures sit near its trunk. A blue shape fills the lower right.” 2. I avoid deciding what an unclear shape represents. “An oval object with three legs” is more faithful than “a rocket” unless the child confirms it. If possible, I ask the child what is happening, since a strange proportion or colour may have a specific story behind it. 3. I separate essentials from flexible details in two short lists. Essentials might include the yellow tree, oversized rabbit, horse’s pose, characters’ arrangement, and red object in the corner. Flexible details might include lighting, textures, materials, environmental detail, and depth. 4. I make clear that the original is the structural source, not just a mood reference. This gives the model room to work while defining the boundaries. 5. I choose a target style, such as a storybook illustration, watercolour, digital painting, comic panel, or 3D scene. 6. I explicitly ask the model to preserve unusual proportions, unexpected colours, expressions, and relationships between characters. I identify details that must remain, even if they appear anatomically impossible or visually awkward. I avoid vague instructions such as “make it better,” because they leave the model to decide what counts as improvement. 7. I use a starting prompt such as: “Use this drawing as the structural blueprint for a storybook illustration. Preserve the composition, relative sizes, poses, expressions and original colour choices. Keep the yellow tree and the rabbit larger than the person. Do not normalize unusual anatomy or remove unexplained objects. Add texture, lighting and environmental depth around the existing scene.” 8. I treat comparison as part of the workflow because specific instructions do not guarantee a faithful result. I expand the world around the invention: a blank background can become a forest, a blue patch can become a pond if that interpretation fits, and two skiing figures can gain a mountain landscape. 9. I keep the original spatial relationships: who stands beside whom, what dominates the scene, which way characters face, and where empty space matters. I give colour the same care, so a yellow tree does not quietly become green because the model expects green trees. 10. I place the source and result side by side and check whether a character disappeared, the rabbit shrank, the palette changed, or the model invented a focal point, “fixed” an unusual shape, or altered the story. I also watch for generic cuteness—big eyes, pastel colours, and perfect symmetry—which can make very different drawings converge into the same image. I ask whether the result still feels like an expansion of this exact drawing. 11. If most of the result works, I request a targeted edit, such as: “Keep this version, but restore the original hair shape.” “Make the rabbit larger again.” “Return the tree to yellow.” 12. I check the whole image after each local edit because changes can affect other areas. I keep useful versions instead of repeatedly starting over. 13. I treat handwritten text separately. I do not guess unclear wording, correct spelling, or invent labels. If the tool cannot reproduce the text faithfully, I retain the original lettering through a separate editing step. 14. I save the original photo or scan alongside the result, instructions, and useful intermediate versions. I also keep a brief note about protected details to make future transformations easier. 15. Once I have a faithful version, I explore alternatives such as a space setting, surreal painting, or cinematic scene, keeping them as branches from the original. Across several drawings, I preserve recurring colours, shapes, and characters without forcing every image into one rigid style. The child has already decided who exists, what matters, how large things are, and what belongs together. Those decisions are creative direction. A drawing does not need to become less itself to become more finished. The impossible colour may be the atmosphere, and the awkward shape may be the character. The question I return to is: “How can I help this drawing become more fully what it already is?”

AI image editors often change more than requested. Removing an object can alter the lighting, cleaning up a background can redraw a face, and fixing one detail can shift typography or proportions. That may be useful for creative transformation, but it is a problem when the goal is to change one thing and leave everything else alone. I now treat precision-editing prompts as preservation contracts. A precise edit needs two instructions: what must change and what must remain untouched. Failures happen when the requested change is clear but the boundaries are vague. For example, “Remove the grey area between the neck and ponytail” may also trigger a redrawn neck, shifted hair, or a smoothed silhouette. The result may look polished, but it is no longer faithful. Step-by-step: 1. I identify the smallest possible edit region. Instead of saying, “Clean up the portrait,” I specify, “Change only the grey area between the neck and ponytail to match the white background.” A narrow target limits unintended reinterpretation. 2. I name the protected elements instead of relying on “Leave everything else alone.” I identify faces, hands, text, positions, proportions, colors, silhouettes, and framing, and include: “Do not add, remove or redesign unrelated elements. Do not make compensatory changes elsewhere.” Otherwise, the AI may adjust shadows, symmetry, or composition to “balance” the edit. 3. I lock all text. Image models may rewrite text they were never asked to touch, so I specify: “Do not regenerate, correct, restyle or replace any text. Preserve spelling, capitalization, punctuation, line breaks, font appearance, size and placement exactly.” I protect labels, handwriting, and captions too. 4. I preserve geometry and framing. For a local correction, I specify: “Preserve all original proportions, positions and geometry outside the edit region. Keep the exact canvas size, aspect ratio and crop. Do not reframe or extend the image.” This is crucial for diagrams, collages, and layouts where spatial relationships matter. 5. I explain the background logic. For example: “Remove the object and reconstruct only the background naturally behind it, continuing the existing pattern without new texture or shadows.” White should stay white, patterns should continue, and shadows should follow the existing light. Removal is not redesign. 6. I ask the model to match the original style, including texture, grain, lighting, line quality, colors, and level of detail. Hand-drawn marks should not become glossy, and grainy photographs should not become plastic-smooth. I preserve imperfections such as crooked lines, collage edges, and photographic wear when they are meaningful. 7. I make fidelity the priority by stating: “This is a local corrective edit, not a reinterpretation or restyle. Prioritize fidelity to the source over beautification.” For risky features, I prohibit redrawn faces, altered text, new backgrounds, cropping, and unrelated corrections. If the target is ambiguous, I add: “If uncertain, preserve the original rather than editing additional areas.” 8. I make delicate changes separately. When there are multiple corrections, I edit one thing, inspect it, and then continue. More requests create more room for drift. 9. I verify what did not change by comparing the result with the source. I check the text, edges, faces, proportions, placement, crop, colors, and background, then ask two questions: Did the requested change happen? What else changed? The second question reveals hidden failures. I use this reusable prompt structure: Change: [specific requested change] Edit only: [exact region or element] Preserve exactly: [text, objects, positions, proportions, colors, framing, etc.] Do not: [likely unwanted changes] Style: Match the existing texture, lighting, grain, line quality and detail. Priority: Fidelity over beautification. If uncertain, preserve the source. For example, instead of writing, “Remove the grey bit under the hair,” I write: “Change only the grey area between the neck and ponytail to the same white as the surrounding background. Preserve the neck, chin, ponytail, silhouette, proportions, crop and all other colors. Do not redraw or beautify the face, hair or body. Do not alter anything outside the target region. Match the existing flat visual style.” The overall workflow is: define the smallest edit, identify protected content, preserve geometry, style, and background logic, forbid unrelated changes, prioritize fidelity, and compare the result against the source. The edit instruction tells the model where to go. The preservation constraints tell it where not to wander. The best corrective edit is not the most imaginative; it changes the target while everything else remains stubbornly itself.

My friend shot a short film and gave me his hard drive to help with post-production. It contained more than 400 clips, with no clear way to tell when each one had been shot. I asked Claude to review the metadata for every clip and group them by timestamp, so they were organized by shoot session instead of sitting in one large folder. Claude also built a spreadsheet listing each clip, its shoot session, the date, runtime, and file size. This saved my friend the hassle of reviewing every clip manually. Step-by-step: 1. I received my friend’s hard drive with more than 400 short-film clips. 2. I asked Claude to review the metadata for every clip. 3. I had Claude group the clips by timestamp and organize them by shoot session. 4. I had Claude create a spreadsheet with each clip’s shoot session, date, runtime, and file size. 5. My friend could avoid going through every clip manually.

I'm a British Academy (BAFTA) voting member. Every autumn, the members' screener platform fills up with more than 100 eligible films—major contenders, obscure documentaries, and hidden gems all mixed together. I have a few weeks to decide what deserves my viewing time before I vote. The fun challenge is finding the sleeper nobody has heard of yet; the problem is the sheer volume of clutter I have to sift through. I built a workflow in Claude Cowork that turns the overwhelming list into an ongoing conversation with a curator that knows my taste. I started by inputting my film notes from previous years: what I watched each season, what I loved, what I abandoned, and why. Claude stores this in persistent memory alongside a growing profile of my taste, including my go-to critic, my low tolerance for slow cinema, the kinds of films that reward me, and my rule that a film gets minutes—not acts—to earn my time. Every session starts with this context already loaded. When a new batch of films drops, Claude opens the screener platform in its built-in browser, reads the full slate, and sends research agents to cover every title in parallel. They research the director, cast, runtime, Rotten Tomatoes scores, festival prizes, awards buzz, and whether my favourite critic has reviewed each film. Claude even pulls the transcripts of his YouTube reviews and summarises the verdict. In about 20 minutes, a 37-film batch is triaged into three categories: priority viewing, worth sampling, and skips. Slow starters are flagged so I know which films need a committed evening and which get 10 minutes to prove themselves. As I watch, I feed verdicts back in plain English, such as “gave it ten minutes, too abstract.” Claude updates both the tracker and its model of my taste, so each recommendation round becomes sharper. By voting time, I have a complete, searchable record of everything I watched and what I thought of it, built conversationally throughout the season. Step-by-step: 1. I input my film notes from previous years—everything I watched, loved, and abandoned, along with my reasons. Claude saves them to persistent memory, which every future session loads automatically, and builds a profile of my taste from them. 2. When new films appear on the members' platform, Claude opens it in its built-in browser and reads the full list of titles. 3. Claude launches parallel research agents to cover every title, including the cast, crew, runtime, Rotten Tomatoes scores, festival prizes, awards buzz, and my preferred critic's verdict from YouTube review transcripts. 4. Claude compiles a triage tracker that ranks every film by viewing priority, flags likely sleepers, and warns me about slow starters. 5. As I sample films, I report quick verdicts in plain English. Claude logs them and continuously refines its profile of my taste, so recommendations improve throughout the season. 6. At voting time, I have a complete record of what I watched, what I thought, and why, built conversationally across the season.

I have tried many AI Tools however - I feel the time I Input data and effort into schedulign and resourcing is the problem - I need a Fix or a rubust reliable AI workflow so i can buy my time back. Any suggestions?

GeneTrust AI — Asistente Genético Virtual (AGV) es una plataforma tecnológica concebida bajo una arquitectura basada en evidencia (Evidence-First), asistida por IA y gobernada por humanos. Está diseñada para facilitar el acceso, la estructuración, la exploración, la clasificación, la priorización y la síntesis trazable de evidencia genética y biomédica. Su diseño busca prevenir riesgos críticos de la IA en biomedicina, como las alucinaciones, las referencias no resolubles, la pérdida de procedencia, el ocultamiento de contradicciones y la atribución indebida de autoridad clínica. Aplica operaciones aditivas y no destructivas: filtrar, clasificar o priorizar los datos nunca modifica ni reescribe la evidencia fuente original. Además, preserva la ambigüedad y los elementos no clasificables en lugar de forzar conclusiones artificiales. El stack tecnológico incluye React, TypeScript, Node.js y servicios en la nube. La plataforma integra un enfoque de múltiples proveedores y agentes de IA, con modelos generativos desacoplados, junto con una clasificación multidimensional de la evidencia mediante la abstracción de un Structured Decision Provider (SDP, Proveedor de Decisión Estructurada). Utiliza específicamente TypeSafe AI, con su modelo Jev, como proveedor de decisiones estructuradas bajo estrictas políticas de activación y gobernanza por dimensión. La IA y sus agentes asisten en la organización, exploración y síntesis de información compleja, pero no actúan como una autoridad diagnóstica ni determinan por sí mismos la verdad clínica. Sus salidas se manejan como inferencias computacionales controladas, separadas estrictamente de la evidencia fuente y sujetas a validación humana explícita. Step-by-step: 1. Accedo y estructuro la evidencia genética y biomédica dentro de una arquitectura basada en evidencia. 2. Exploro, clasifico y priorizo la información mediante operaciones aditivas y no destructivas, sin modificar ni reescribir la evidencia fuente original. 3. Mantengo la procedencia, las contradicciones, la ambigüedad y los elementos no clasificables, evitando forzar conclusiones artificiales. 4. Utilizo múltiples proveedores y agentes de IA, con modelos generativos desacoplados, para organizar, explorar y sintetizar la información compleja. 5. Aplico una clasificación multidimensional mediante la abstracción de un Structured Decision Provider (SDP), utilizando TypeSafe AI y su modelo Jev bajo políticas estrictas de activación y gobernanza por dimensión. 6. Trato las salidas de la IA como inferencias computacionales controladas, las separo de la evidencia fuente y requiero una validación humana explícita antes de utilizarlas.

Maintaining an AI-built property finance app: proven Codex testing, release and handoff workflows? I’m building a small family property-management app using ChatGPT for requirements and Codex desktop on Linux for implementation, with TypeScript, PostgreSQL/Supabase and Vercel. It covers renovation forecasts, quotes, costs and payments, operating income and expenses, and family loans. The app is already in use. We have automated checks, isolated synthetic data, financial reconciliation and controlled releases. I’m looking for practical experience from people who have maintained an AI-built financial or admin app beyond its prototype: 1. How do you keep requirements, acceptance tests, candidate versions and deployed releases aligned across long Codex threads and workspaces? 2. What browser-testing and release-evidence workflow has caught failures that builds and unit tests missed? 3. Which GitHub skills or agent packages have improved your results without conflicting instructions or excessive process? 4. What works well for translating approved mockup images into maintainable application code, including screenshot comparison and mobile behaviour? 5. What broke after deployment, and what did you change? Links to maintained repositories, runbooks, example acceptance records or fictional-data screen recordings would be especially useful. The aim is to improve the existing application and workflow while keeping everyday tasks understandable for non-technical family users.

The content, business plan, and trajectory is established. Coding and execution is the precise matter that is required. I lack coding skills, and have yet to acquire a proper coder to fit the context.

A chocolate factory client had a simple but expensive sales problem: good conversations with buyers sometimes went quiet because nobody followed up at the right time. Some opportunities were waiting on sample feedback, pricing, private-label details, minimum order quantities (MOQs), packaging, or distributor discussions. The opportunities were still active, but they were easy to lose track of. I built a follow-up workflow for the client in Awish. Step-by-step: 1. I opened the Awish chat and wrote: “Track our open sales opportunities from Microsoft Teams and Google Sheets. Every business day, review the latest customer conversation, sales stage, planned follow-up date, customer interest, and how long we’ve been out of touch. Find opportunities that need follow-up around samples, quotations, private label, MOQ, packaging, or distribution. Explain why each customer should be contacted, recommend the next action, and prepare a personalized Teams follow-up message based on the real conversation history. Never send anything without the salesperson’s approval. When the customer replies, close the old follow-up action and update the Sales Master record.” 2. Awish understood the request, planned the workflow, and selected Microsoft Teams and Google Sheets for the process. 3. I connected the client’s accounts, reviewed the plan, and approved the automation. 4. When a new customer conversation takes place in Teams, Awish updates the related sales record. 5. Every business day, it reviews open opportunities and finds deals that are overdue, at risk, or ready for the next follow-up. 6. For each opportunity, it explains why follow-up is needed and what the salesperson should do next. 7. Awish prepares a personalized Teams message using the actual conversation history, but waits for the salesperson’s approval before sending anything. 8. When the customer responds, the previous follow-up is closed and the opportunity status in Google Sheets is updated automatically. Now the sales team gets a prioritized daily list of who needs attention, why they need attention, and what the next message should be—without manually reviewing every old conversation. I want to keep building more workflows like this for real businesses. If you have a complex, repetitive process in your company that you think should be automated but you’re not sure how to build it, send it to me. I’d be happy to see if I can turn it into a working automation with Awish.

I used Claude to review the book *Why Machines Learn*. Then I went through each chapter while Claude questioned me at relevant points. To make the concepts more concrete, I used interactive graphs with x and y axes. The first graph shows the relationship between x and a function, along with the slope of the curve. It uses f(x) = x³ with a tangent line illustrating the derivative. I can drag a slider to move the point and watch the slope change, then switch to 3x to compare it with a straight line. The graph demonstrates several ideas: - Sliding x to 0 on x³ makes the tangent flat and the slope 0, so a small nudge barely changes the output. This is similar to a flat spot at the ends of a sigmoid, where learning stalls. - Sliding to 2 and then to −2 shows a slope of 12 at both points. Because the derivative is 3x², it is positive on both sides and the curve is always climbing. - Switching to 3x makes the tangent lie directly on top of the line, with a constant slope of 3 everywhere. - The nudge-check box tracks the derivative closely, showing that the derivative is essentially “change in output ÷ small change in input.” The second graph shows the input and output layers of a small pattern detector. It represents a 2 × 2 image flattened into four inputs, a hidden layer of two neurons, and two outputs. The weights are set by hand so the network can distinguish a vertical line from a horizontal one. I can click the pixels to change the image and watch the numbers flow through the network. The setup uses weight +1, weight −1, ReLU activation defined as max(0, z), softmax output, and all biases set to 0. For example: - Flatten: x = [1, 0, 1, 0] - h1 = ReLU((1)(1) + (-1)(0) + (1)(1) + (-1)(0)) = ReLU(2) = 2 — left column versus right column - h2 = ReLU((1)(1) + (1)(0) + (-1)(1) + (-1)(0)) = ReLU(0) = 0 — top row versus bottom row - Output: scores (2, 0) → softmax → Vertical 88%, Horizontal 12% The graph includes examples for a left vertical line, top horizontal line, right vertical line, and diagonal. Seeing the numbers change as the factors change really helps bring these concepts to life for me. Step-by-step: 1. I used Claude to review *Why Machines Learn* and question me as I worked through each chapter. 2. I used an interactive x-and-y graph to explore how a function changes and how its slope relates to the derivative. 3. I moved the slider across f(x) = x³, including 0, 2, and −2, to observe how the tangent and slope changed. 4. I switched the graph to 3x and used the nudge-check box to compare a constant derivative with the changing derivative of x³. 5. I explored a second interactive diagram representing a 2 × 2 image, four inputs, two hidden neurons, and two outputs. 6. I clicked different pixels and watched the ReLU activations, output scores, and softmax predictions change for vertical, horizontal, and diagonal patterns.

I write science fiction and use ChatGPT to make sure my stories stay within certain rules. I use Willow Voice AI for dictation and video generators to create proof-of-concept visuals. I also recently used AI to find quick answers within a 40-minute YouTube video. I turned it from watchbait into quick answers. Step-by-step: 1. I use ChatGPT to check that my science-fiction stories stay within certain rules. 2. I use Willow Voice AI to dictate my writing. 3. I use video generators to create proof-of-concept visuals. 4. I use AI to find quick answers within a 40-minute YouTube video. 5. I turn the video from watchbait into quick answers.

Most people remember the predictions they got right and rationalize the ones they got wrong. This workflow forces every meaningful forecast into a structured record so you can measure not only whether you were right, but also whether your confidence was justified. The objective is not simply to make more correct predictions. It is to make your stated confidence increasingly match reality. For every forecast, record the following: Step-by-step: 1. State exactly what you believe will happen. The claim must be specific enough to resolve later. Bad: “AI stocks look vulnerable.” Better: “The Nasdaq-100 will fall at least 15% from its current level within six months.” 2. Assign a numeric probability from 0–100%. Avoid vague terms such as “likely” or “probably” without attaching a percentage. 3. Define when the prediction should resolve, such as within 30 days, by year-end, or within 12 months. 4. Identify the one to three factors most responsible for the forecast being correct. This makes the causal thesis explicit. 5. Specify what evidence would materially weaken or invalidate the forecast. This prevents moving the goalposts later. 6. Define in advance what counts as correct, incorrect, or ambiguous. Do not wait until the outcome is known to decide how success will be measured. 7. Freeze the forecast by preserving the original forecast exactly. If new information materially changes your view, create a new forecast rather than editing the old one. This preserves the forecasting record. 8. At the end of the time horizon, record what actually happened. Then distinguish between a thesis error, in which the underlying reasoning was wrong; a timing error, in which the thesis may have been right but the forecast horizon was wrong; and a driver error, in which the expected outcome occurred but for different reasons. 9. Review calibration by comparing probability bands with actual results over time. If forecasts assigned 70% confidence are correct about 70% of the time, your confidence is well calibrated. If high-confidence forecasts fail too often, you are overconfident. If lower-confidence forecasts succeed too often, you may be underconfident. Do not overinterpret tiny sample sizes; calibration becomes more meaningful as forecasts accumulate. 10. Periodically review error patterns, including overconfidence, underconfidence, timing errors, weak assumptions, ignored counterevidence, wrong key drivers, poor resolution criteria, and repeated mistakes by forecasting domain. Use those patterns to improve future forecasts. Copy-and-paste prompt Act as a forecasting and calibration analyst. Whenever I make a meaningful prediction, create a forecast record using: CLAIM: PROBABILITY: TIME HORIZON: KEY DRIVERS: DISCONFIRMING SIGNALS: RESOLUTION CRITERIA: Make the prediction specific enough to evaluate later. Require a numeric probability and a defined time horizon. Preserve the original forecast exactly. If new information materially changes the forecast, create a new version rather than rewriting the original. When the forecast resolves, add: OUTCOME: RESULT: THESIS ACCURACY: TIMING ACCURACY: DRIVER ACCURACY: WHAT WAS RIGHT: WHAT WAS WRONG: CALIBRATION REVIEW: LESSON: Periodically review resolved forecasts for: Accuracy by probability band Overconfidence Underconfidence Timing errors Repeated bad assumptions Strongest forecasting domains Weakest forecasting domains Never rewrite history to make an old forecast look better. The objective is not to maximize correct predictions. The objective is to make confidence increasingly match reality.