AB's Reflections

Tag: thoughts

  • GitHub’s SpecKit: The Structure Vibe Coding Was Missing

    GitHub’s SpecKit: The Structure Vibe Coding Was Missing

    When I first started experimenting with “vibe coding,” building apps with AI agents felt like a superpower. The ability to spin up prototypes in hours was exhilarating. But as I soon discovered, the initial thrill came with an illusion. It was like managing a team of developers with an attrition rate measured in minutes—every new prompt felt like onboarding a fresh hire with no idea what the last one had been working on.

    The productivity boost was real, but the progress was fragile. The core problem was context—a classic case of the law of leaky abstractions applied to AI. Models would forget why they made certain choices or break something they had just built. To cope, I invented makeshift practices: keeping detailed dev context files, enforcing strict version control with frequent commits, and even asking the model to generate “reset prompts” to re-establish continuity. Messy, ad hoc, but necessary.

    That’s why GitHub’s announcement of SpecKit immediately caught my attention. SpecKit is an open-source toolkit for what they call “spec-driven development.” Instead of treating prompts and chat logs as disposable artifacts, it elevates specifications to first-class citizens of the development lifecycle.

    In practice, this means:

    • Specs as Durable Artifacts: Specifications live in Git alongside your code—permanent, version-controlled, and not just throwaway notes.
    • Capturing Intent: They document the why—the constraints, purpose, and expected behavior—so both humans and AI stay aligned.
    • Ensuring Continuity: They serve as the source of truth, keeping projects coherent across sessions and contributors.

    For anyone who has tried scaling vibe coding beyond a demo, this feels like the missing bridge. It brings just enough structure to carry a proof-of-concept into maintainable software.

    And it fits into a larger story. Software engineering has always evolved in waves—structured programming, agile, test-driven development. Each wave added discipline to creativity, redefining roles to reflect new economic realities—a pattern we’re seeing again with agentic coding. Spec-driven development could be the next step:

    • Redefining the Developer’s Role: Less about writing boilerplate, more about designing robust specs that guide AI agents.
    • Harnessing Improvisation: Keeping the creative energy of vibe coding, but channeling it within a coherent framework.
    • Flexible Guardrails: Not rigid top-down rules, but guardrails that allow both creativity and scalability.

    Looking back, my dev context files and commit hygiene were crude precursors to this very idea. GitHub’s SpecKit makes clear that those instincts weren’t just survival hacks—they pointed to where the field is heading.

    The real question now isn’t whether AI can write code—we know it can. The question is: how do we design the frameworks that let humans and AI build together, reliably and at scale?

    Because as powerful as vibe coding feels, it’s only when we bring structure to the improvisation that the music really starts.

    👉 What do you think—will specs become the new lingua franca between humans and AI?

  • India’s Race Between Demography and AI

    India’s Race Between Demography and AI

    Artificial Intelligence is often framed as a threat to jobs—a disruptive force poised to replace human labour at an unprecedented scale. From call centres to accounting firms, from routine coding to legal research, Generative AI and automation tools are already demonstrating capabilities that once seemed untouchable. The fear of widespread job loss is not unfounded. McKinsey, among others, estimates that nearly a quarter of global work activities could be automated by the early 2030s.

    Yet, there is another equally significant demographic trend reshaping the labour market—the aging of populations. In countries such as Japan, South Korea, Germany, and even China, the working-age population is shrinking. This is not because jobs are disappearing, but because people are. Fertility rates have fallen below replacement levels, and the proportion of elderly citizens is rising sharply. These nations face a paradox: they need more workers but have fewer people available to fill roles.

    AI and Aging: Complementary Forces in Developed Countries

    This is where AI and aging unexpectedly complement each other. In economies that are already greying, AI is less a destroyer of jobs and more a replacement for the labour that no longer exists. Japan, for example, has pioneered the use of robotics and AI-driven systems not to replace young workers, but to stand in for absent ones—care robots for the elderly, AI-assisted diagnostics for hospitals short on doctors, and factory automation for industries facing chronic staff shortages.

    In such societies, the fear of AI taking away jobs is muted by the demographic reality that many jobs would otherwise remain unfilled. AI is effectively stepping into the gap created by demographic decline. For them, the challenge is not managing unemployment but managing the technological transition in a way that sustains productivity and care standards as their populations age.

    India’s Young Advantage—and the Ticking Clock

    India, however, tells a different story. The country’s demographic structure is still overwhelmingly young. Nearly two-thirds of Indians are in the working-age bracket, and the median age is around 28—more than a decade younger than China or the U.S. This “demographic dividend” has been hailed as India’s biggest economic advantage for the next 10–15 years. But this window is finite.

    Demographers estimate that by the mid-2030s, India’s working-age population will peak. After that, the proportion of elderly citizens will start rising sharply. By 2046, the elderly are projected to outnumber children under 14. In other words, India’s advantage will begin to fade just as many advanced economies have already entered the post-dividend phase. If India cannot create enough productive jobs during this critical decade, its youth bulge may turn into a liability.

    AI’s Adoption Curve

    The question is: will AI go mainstream while India’s workforce is still young? Current projections suggest that large-scale AI adoption is still 5–15 years away. Today’s Generative AI tools, while impressive, remain in an experimental phase. They lack reliability, governance frameworks, and cost efficiency at scale. Gartner’s hype cycle places most AI technologies in the “Trough of Disillusionment,” meaning that widespread productivity gains will take years to materialize.

    If this trajectory holds, AI’s mainstream integration across sectors like healthcare, education, law, and public administration may not happen until the 2030s—roughly the same time that India’s demographic dividend starts to decline. This sets up an intriguing scenario where India’s labour market transition and AI’s maturity could synchronize.

    Possible Scenarios for India

    1. The Collision Scenario:

    If AI adoption accelerates too quickly, India’s youthful workforce may find itself competing against machines for jobs before the country has built a strong industrial and service base. Sectors such as BPO, customer service, and low-skill IT roles—once the backbone of India’s outsourcing economy—could see rapid automation. Without massive reskilling efforts, unemployment among young Indians could spike even as the global economy demands fewer entry-level jobs.

    2. The Missed Opportunity Scenario:

    Alternatively, if AI adoption lags too far behind—say, beyond 2040—India could enter its aging phase without having reaped the productivity gains AI promises. By then, the country would face the dual pressures of a shrinking workforce and a delayed technological transition. This would mirror some of the struggles seen in late-industrializing economies that missed the manufacturing wave.

    3. The Synchronization Scenario:

    The most optimistic possibility is that AI and India’s demographic transition align productively. Over the next decade, India could use its young workforce to build, train, and scale AI systems, preparing the ground for when labour shortages begin. By the time the aging curve hits in the 2035–2040 period, AI could step in not as a threat, but as a productivity amplifier—automating routine tasks while humans focus on complex, creative, or empathetic roles.

    This requires a proactive strategy: early investment in AI literacy, creation of AI-enabled jobs (rather than job replacement), and building a global service economy where Indians are not just users of AI, but architects of AI solutions.

    The Decisive Decade

    India’s story in the 2030s will be defined by the intersection of two megatrends: a maturing workforce and a maturing technology. Whether this convergence leads to disruption or opportunity depends on choices made now—in education, infrastructure, governance, and industry adoption. The challenge is to ensure that when AI becomes mainstream, India’s workforce is not left behind but is ready to ride the wave. The 2020s are not just a decade of demographic advantage—they are the runway for an AI-driven, post-dividend future.

  • The Economic Reality and the Optimistic Future of Agentic Coding

    The Economic Reality and the Optimistic Future of Agentic Coding

    After a couple of months deep in the trenches of vibe coding with AI agents, I’ve learned this much: scaling from a fun, magical PoC to an enterprise-grade MVP is a completely different game.

    Why Scaling Remains Hard—And Costly

    Getting a prototype out the door? No problem.

    But taking it to something robust, secure, and maintainable? Here’s where today’s AI tools reveal their limits:

    • Maintenance becomes a slog. Once you start patching AI-generated code, hidden dependencies and context loss pile up. Keeping everything working as requirements change feels like chasing gremlins through a maze.
    • Context loss multiplies with scale. As your codebase grows, so do the risks of agents forgetting crucial design choices or breaking things when asked to “improve” features.

    And then there’s the other elephant in the room: costs.

    • The cost scaling isn’t marginal—not like the old days of cloud or Web 2.0. Powerful models chew through tokens and API credits at a rate that surprises even seasoned devs.
    • That $20/month Cursor plan with unlimited auto mode? For hobby projects, it’s a steal. For real business needs, I can see why some queries rack up millions of tokens and would quickly outgrow even the $200 ultra plan.
    • This is why we’re seeing big tech layoffs and restructuring: AI-driven productivity gains aren’t evenly distributed, and the cost curve for the biggest players keeps climbing.

    What the Data Tells Us

    That research paper—Measuring the Impact of Early-2025 AI on Experienced Open-Source Developer Productivity—had a surprising conclusion:

    Not only did experienced developers see no time savings on real-world coding tasks with AI, but costs increased as they spent more time reviewing, correcting, and adapting agent output.

    The lesson:

    AI shifts where the work happens—it doesn’t always reduce it. For now, scaling with agents is only as good as your processes for context, review, and cost control.

    Why I Remain Optimistic

    Despite the challenges, I’m genuinely excited for what’s coming next.

    • The platforms and models are evolving at warp speed. Many of the headaches I face today—context loss, doc gaps, cost blind spots—will get solved just as software engineering best practices eventually became codified in our tools and frameworks.
    • Agentic coding will find its place. It might not fully automate developer roles, but it will reshape teams: more focus on high-leverage decisions, design, and creative problem-solving, less on boilerplate and “busy work.”

    And if you care about the craft, the opportunity is real:

    • Devs who learn to manage, review, and direct agents will be in demand.
    • Organizations that figure out how to blend agentic workflows with human expertise and robust process will win big.

    Open Questions for the Future

    • Will AI agentic coding mean smaller, nimbler teams—or simply more ambitious projects for the same headcount?
    • How will the developer role evolve when so much code is “synthesized,” not hand-crafted?
    • What new best practices, cost controls, and team rituals will we invent as agentic coding matures?

    Final thought:

    The future won’t be a return to “pure code” or a total AI handoff. It’ll be a blend—one that rewards curiosity, resilience, and the willingness to keep learning.

    Where do you see your work—and your team—in this new landscape?

  • The Law of Leaky Abstractions & the Unexpected Slowdown

    The Law of Leaky Abstractions & the Unexpected Slowdown

    If the first rush of agentic/vibe coding feels like having a team of superhuman developers, the second phase is a reality check—one that every software builder and AI enthusiast needs to understand.

    Why “Vibe Coding” Alone Can’t Scale

    The further I got into building real-world prototypes with AI agents, the clearer it became: Joel Spolsky’s law of leaky abstractions is alive and well.

    You can’t just vibe code your way to a robust app—because underneath the magic, the cracks start to show fast. AI-generated coding is an abstraction, and like all abstractions, it leaks. When it leaks, you need to know what’s really happening underneath.

    My Experience: Hallucinations, Context Loss, and Broken Promises

    I lost count of the times an agent “forgot” what I was trying to do, changed underlying logic mid-stream, or hallucinated code that simply didn’t run. Sometimes it wrote beautiful test suites and then… broke the underlying logic with a “fix” I never asked for. It was like having a junior developer who could code at blazing speed—but with almost no institutional memory or sense for what mattered.

    The “context elephant” is real. As sessions get longer, agents lose track of goals and start generating output that’s more confusing than helpful. That’s why my own best practices quickly became non-negotiable:

    • Frequent commits and clear commit messages
    • Dev context files to anchor each session
    • Separate dev/QA/prod environments to avoid catastrophic rollbacks (especially with database changes)

    What the Research Shows: AI Can Actually Slow Down Experienced Devs

    Here’s the kicker—my frustration isn’t unique.

    A recent research paper, Measuring the Impact of Early-2025 AI on Experienced Open-Source Developer Productivity, found that experienced developers actually worked slower with AI on real-world tasks. That’s right—AI tools didn’t just fail to deliver the expected productivity boost, they created friction.

    Why?

    • Only about 44% of AI-generated code was accepted
    • Developers lost time reviewing, debugging, and correcting “bad” generations
    • Context loss and reliability issues forced more manual intervention, not less

    This matches my experience exactly. For all the hype, these tools introduce new bottlenecks—especially if you’re expecting them to “just work” out of the box.

    Lessons from the Frontlines (and from Agent Week)

    I’m not alone. In the article What I Learned Trying Seven Coding Agents, Timothy B. Lee finds similar headaches:

    • Agents get stuck
    • Complex tasks routinely stump even the best models
    • Human-in-the-loop review isn’t going anywhere

    But the tools are still useful—they’re not a dead end. You just need to treat them like a constantly rotating team of interns, not fully autonomous engineers.

    Best Practices: How to Keep AI Agents Under Control

    So how do you avoid the worst pitfalls?

    The answer is surprisingly old-school:

    • Human supervision for every critical change
    • Sandboxing and least privilege for agent actions
    • Version control and regular context refreshers

    Again, Lee’s article Keeping AI agents under control doesn’t seem very hard nails it:

    Classic engineering controls—proven in decades of team-based software—work just as well for AI. “Doomer” fears are overblown, but so is the hype about autonomy.

    Conclusion: The Hidden Cost of Abstraction

    Vibe coding with agents is like riding a rocket with no seatbelt—exhilarating, but you’ll need to learn to steer, brake, and fix things mid-flight.

    If you ignore the leaky abstractions, you’ll pay the price in lost time, broken prototypes, and hidden tech debt.

    But with the right mix of skepticism and software discipline, you can harness the magic and avoid the mess.

    In my next post, I’ll zoom out to the economics—where cost, scaling, and the future of developer work come into play.

    To be continued…

  • The Thrill and the Illusion of AI Agentic Coding

    The Thrill and the Illusion of AI Agentic Coding

    A few months ago, I stumbled into what felt like a superpower: building fully functional enterprise prototypes using nothing but vibe coding and AI agent tools like Cursor and Claude. The pace was intoxicating—I could spin up a PoC in days instead of weeks, crank out documentation and test suites, and automate all the boring stuff I used to dread.

    But here’s the secret I discovered: working with these AI agents isn’t like managing a team of brilliant, reliable developers. It’s more like leading a software team with a sky-high attrition rate and non-existent knowledge transfer practices. Imagine onboarding a fresh dev every couple of hours, only to have them forget what happened yesterday and misinterpret your requirements—over and over again. That’s vibe coding with agents.

    The Early Magic

    When it works, it really works. I’ve built multiple PoCs this way—each one a small experiment, delivered at a speed I never thought possible. The agents are fantastic for “greenfield” tasks: setting up skeleton apps, generating sample datasets, and creating exhaustive test suites with a few prompts. They can even whip up pages of API docs and help document internal workflows with impressive speed.

    It’s not just me. Thomas Ptacek’s piece “My AI Skeptic Friends Are All Nuts” hits the nail on the head: AI is raising the floor for software development. The boring, repetitive coding work—the scaffolding, the CRUD operations, the endless boilerplate—gets handled in minutes, letting me focus on the interesting edge cases or higher-level product thinking. As they put it, “AI is a game-changer for the drudge work,” and I’ve found this to be 100% true.

    The Fragility Behind the Hype

    But here’s where the illusion comes in. Even with this boost, the experience is a long way from plug-and-play engineering. These AI coding agents don’t retain context well; they can hallucinate requirements, generate code that fails silently, or simply ignore crucial business logic because the conversation moved too fast. The “high-attrition, low-knowledge-transfer team” analogy isn’t just a joke—it’s my daily reality. I’m often forced to stop and rebuild context from scratch, re-explain core concepts, and review every change with a skeptical eye.

    Version control quickly became my lifeline. Frequent commits, detailed commit messages, and an obsessive approach to saving state are my insurance policy against the chaos that sometimes erupts. The magic is real, but it’s brittle: a PoC can go from “looks good” to “completely broken” in a couple of prompts if you’re not careful.

    Superpowers—With Limits

    If you’re a founder, product manager, or even an experienced developer, these tools can absolutely supercharge your output. But don’t believe the hype about “no-code” or “auto-code” replacing foundational knowledge. If you don’t understand software basics—version control, debugging, the structure of a modern web app—you’ll quickly hit walls that feel like magic turning to madness.

    Still, I’m optimistic. The productivity gains are real, and the thrill of seeing a new prototype come to life in a weekend is hard to beat. But the more I use these tools, the more I appreciate the fundamentals that have always mattered in software—and why, in the next post, I’ll talk about the unavoidable reality check that comes when abstractions leak and AI doesn’t quite deliver on its promise.

    To be continued…

  • Shadow AI, Friction Fatigue & the Flexibility Gap: 5 Lessons from the Ivanti Tech at Work 2025 Report

    Shadow AI, Friction Fatigue & the Flexibility Gap: 5 Lessons from the Ivanti Tech at Work 2025 Report

    The Ivanti Tech at Work 2025 report isn’t just a workplace tech survey — it’s a mirror to how modern organizations are struggling (and sometimes succeeding) to adapt to the realities of hybrid work, AI adoption, and employee expectations.

    Here are 5 insights that stood out — and why they matter for anyone building teams, tools, or trust in the modern workplace.

    🔗 Read the full report

    1. Shadow AI Is a Trust Problem, Not Just a Tech One

    Nearly 1 in 3 workers admit they use AI tools like ChatGPT in secret at work.

    Why the secrecy?

    According to the report:

    • 36% want a competitive edge
    • 30% fear job cuts or extra scrutiny
    • 30% say there’s no clear AI usage policy
    • 27% don’t want their abilities questioned

    This is more than a governance issue. It’s a cultural signal.
    Employees are turning to AI to be more productive — but doing so under the radar signals a trust deficit and policy vacuum.

    💡 What to do:
    Leaders need to replace silence with structure — with clear, enabling policies that promote responsible AI use, and an environment where value creation matters more than screen time.

    2. The Flexibility Paradox: High Demand, Low Supply

    83% of IT professionals and 73% of office workers value flexibility highly — but only ~25% say they actually have it.

    Even as companies trumpet hybrid work, asynchronous enablement, autonomy, and outcome-based work norms haven’t caught up. The result? Disengagement and frustration.

    💡 What to do:
    Revisit what flexibility really means. It’s not just about where people work — it’s how they work.
    That means:

    • Tools for async collaboration
    • Decision-making frameworks for remote teams
    • Leaders modeling flexible behaviors

    3. Presenteeism Is the New Fatigue

    The report highlights “digital presenteeism”: workers pretending to be active — jiggling mice, logging in early — to appear productive.

    • 48% say they dislike their job but stay
    • 37% admit to showing up without doing meaningful work

    These are signs of unclear expectations and poor workflow design — not disengagement alone.

    💡 What to do:
    Audit for friction, not just lag.
    Look at your workflows, KPIs, and culture. Are people forced to perform busyness instead of real value?

    4. The Digital Experience Gap Is Real

    While flexible work is valued, many workers find it harder to work outside the office. The report notes:

    • 44–49% say collaboration is easier in-office
    • 36–48% say manager access is better in-office
    • 16–26% say apps are easier to use from the office

    💡 What to do:
    Enable remote-first experience, not just policy:

    • Seamless access to tools and systems
    • Integrated collaboration platforms
    • AI-powered support and IT workflows

    5. Redesign for Trust, Not Just Tools

    The big takeaway?

    Workers don’t just need better AI — they need clarity on what’s allowed
    They don’t just need more flexibility — they need workflows that enable it
    They don’t just need faster tools — they need a culture that values trust over control

    Final Thoughts

    The Ivanti Tech at Work 2025 report is a diagnostic — revealing what happens when new tools are bolted onto outdated operating models.

    For leaders, the message is clear:

    We need to evolve not just our tech stack, but our trust stack.

    🔗 Read the full report

  • From Productivity to Progress: What the New MIT-Stanford AI Study Really Tells Us About the Future of Work

    From Productivity to Progress: What the New MIT-Stanford AI Study Really Tells Us About the Future of Work

    A new study from MIT and Stanford just rewrote the AI-in-the-workplace narrative.

    Published in Fortune this week, the research shows that generative AI tools — specifically chatbots — are not only boosting productivity by up to 14%, but they’re also raising earnings without reducing work hours.

    “Rather than displacing workers, AI adoption led to higher earnings, especially for lower-performing employees.”

    Let that sink in.

    🧠 AI as a Floor-Raiser, Not a Ceiling-Breaker

    The most surprising finding?
    AI’s greatest impact was seen not among the top performers, but among lower-skilled or newer workers.

    In customer service teams, the AI tools essentially became real-time coaches — suggesting responses, guiding tone, and summarizing queries. The result: a productivity uplift and quality improvement that evened out performance levels across the team.

    This is a quiet revolution in workforce design.

    In many traditional orgs, productivity initiatives often widen the gap between high and average performers. But with AI augmentation, we’re seeing the inverse — a democratization of capability.

    💼 What This Means for Enterprise Leaders

    This research confirms a pattern I’ve observed firsthand in consulting:
    The impact of AI is not just technical, it’s organizational.

    To translate AI gains into business value, leaders need to:

    ✅ 1. Shift from Efficiency to Enablement

    Don’t chase cost-cutting alone. Use AI to empower more team members to operate at higher skill levels.

    ✅ 2. Invest in Workflow Design

    Tool adoption isn’t enough. Embed AI into daily rituals — response writing, research, meeting prep — where the marginal gains accumulate.

    ✅ 3. Reframe KPIs

    Move beyond “time saved” metrics. Start tracking value added — better resolutions, improved CSAT, faster ramp-up for new hires.

    🔄 A Playbook for Augmented Teams

    From piloting GPT agents to reimagining onboarding flows, I’ve worked with startups and enterprise teams navigating this shift. The ones who succeed typically follow this arc:

    1. Pilot AI in a high-volume, low-risk function
    2. Co-create use cases with users (not for them)
    3. Build layered systems: AI support + human escalation
    4. Train managers to interpret, not just supervise, AI-led work
    5. Feed learnings back into process improvement loops

    🔚 Not AI vs Jobs. AI Plus Better Jobs.

    The real story here isn’t about productivity stats. It’s about potential unlocked.

    AI is no longer a futuristic experiment. It’s a present-day differentiator — especially for teams willing to rethink how work gets done.

    As leaders, we now face a simple choice:

    Will we augment the talent we have, or continue to chase the talent we can’t find?

    Your answer will shape the next 3 years of your business.

    🔗 Read the original article here:

    Fortune: AI chatbots boost earnings and hours, not job loss

    Want to go deeper? I’m working on a new AI augmentation playbook — DM me or sign up for updates.

    #AI #FutureOfWork #EnterpriseStrategy #GTM #DigitalTransformation #Chatbots #Productivity #ConsultingInsights

  • Decoding Pharma Analytics: Customer Segmentation and Engagement Strategies

    Decoding Pharma Analytics: Customer Segmentation and Engagement Strategies

    I recently had the opportunity to share a fascinating online lecture to a group of MBA students about Pharmaceutical Analytics, specifically focusing on customer segmentation and engagement strategies.

    I have created a summary podcast using presentation, sources and audio recording from the session which you can check out below:

    Also sharing the presentation from the session.

    Building Real-World Models lecture 2025.01.14Download

    Do get in touch if you would like me to hold similar sessions for your organization or institution.

  • The Moon in 2024 through the S23 & more

    The Moon in 2024 through the S23 & more

    It’s been a while since I posted about my photography activities and the phone reviews, and what better time than the new year to look back at some of the memories created over the last year.

    Here’s a couple of moonshots taken last year with my current phone – the Samsung Galaxy S23 Ultra that I got back in 2023. I’ve put the 30x zoom versions first followed by the regular wide angle photo to give a sense of the zoom. All were taken during our 2024 summer trip to Kashmir (first set in Phalagam, second in Srinagar on the Dal Lake).

    While the zoomed in images may not be the sharpest, the S23 Ultra has finally made me stop missing my DSLR with its zoom lenses. There may have been a lot of controversy over the artificial enhancement in the moon photos taken with Samsung phones, but I have been really pleased with the performance of the 10x lens which enables shots that were unthinkable on mobile phones.

    The moon is of course not the only object that you can shoot, so here’s a few more including the Sun, a rainbow and more travel photos.

    Do follow me on Instagram for more.

  • Dubai Diaries: Running LLMs & Stable Diffusion locally on a gaming laptop

    Dubai Diaries: Running LLMs & Stable Diffusion locally on a gaming laptop

    I previously wrote about the second device that I got about coming to Dubai, but not much about the first one which was a gaming laptop. So here’s a bit about the laptop which also doubles as a local AI driver thanks to the Nvidia GPU (the RTX3060).

    Soon after getting it back in 2022, I tried running the Stable Diffusion models and it was quite a bit of an upgrade over my original attempt on a plain GPU-less Windows machine. The generation times came down to 10s or so, and has gotten even faster as the models and tools have been optimised over the last couple of years. There are quite a few projects available on GitHub if you want give it a try – AUTOMATIC1111 and easydiffusion are among the more popular options. Nvidia has also got a TensorRT extension to further improve performance.

    With that out of the way, I also discovered LM Studio which allows you to run LLMs locally with a chat like interface thrown in, and you can access a bunch of models like Meta’s LLama. The response times are of course not as fast as the freely available online options like ChatGPT, Claude, Gemini and the likes, but you effectively get unlimited access to the model.

    Here’s an example from a conversation I had with LLama regarding the coffee meme from Ace Attorney game series: