Building autonomous AI agents. Notes from the team building Gravity. AI workflows, the future of recurring work, and what we learn along the way.
An agent stopped mid-task is not the same as an agent that finished. If the agent has already taken some actions and not others, the world is in an inconsistent state. A flight booked, a hotel not yet booked. A label…
Read post →
Setting up a first AI agent is straightforward in 2026. The platforms work, the model quality is high enough, and the integrations cover most things a small business or solo professional cares about. What is not…
Read post →
"Roll back the agent's action" is not one operation. It is four operations, each appropriate for a different tier of action. Treating all rollbacks as equal leads to over-engineered Tier 1 actions and…
Read post →
Code fails loudly. An exception is thrown, the process crashes, the error reaches the dashboard, the operator gets an alert. AI agents fail differently. The agent picks a wrong interpretation, takes a wrong action,…
Read post →
Zapier and AI agents solve overlapping problems with different mental models. Zapier is trigger-step-action: a fixed pipeline that runs the same way every time. AI agents are outcome-driven: the agent picks the…
Read post →
An agent that fails small is a fixable problem. An agent that fails catastrophically is a customer-trust event, a financial loss, or both. The difference is blast-radius control: the set of limits that bound what the…
Read post →
Email is the master credential. Most accounts route password resets through it; many financial accounts treat email-confirmed-this as proof of identity. Granting an AI agent access to email is the most consequential…
Read post →
Per-task token cost is the cheapest, easiest, and most misleading number you can quote when estimating an agent's deployed cost. The number is right for a single successful invocation in development. It is wrong,…
Read post →
Agent failures look like bugs but behave differently. A traditional code bug is deterministic; the same input always produces the same wrong output. An agent failure is a decision the agent made on an input where…
Read post →
An AI agent is not magic, even when it acts like it. The mechanism underneath is straightforward enough that a non-developer can hold the whole picture in their head. This explainer walks through the picture without…
Read post →
This is the post I avoided writing for a year. The financial postmortem at the honest cost of three shutdowns was uncomfortable. This one is harder. Three startup shutdowns over four years did real damage to my…
Read post →
Workflow automation and AI agents look like the same product category from a distance. They run things on your behalf, integrate with your tools, and remove drudgery. Up close, they are different abstractions, with…
Read post →
Sales conversations in 2025 and 2026 routinely include the phrase "AI agents are basically RPA 2.0". The phrase is a category error. Agents and RPA solve overlapping problems but at different abstraction layers, with…
Read post →
"Copilot" and "agent" are now used interchangeably in marketing copy, which makes the buyer's job harder. The two categories are different. A copilot helps a person do their work; an agent does the work and reports…
Read post →
The terms "AI agent", "chatbot", and "assistant" are used interchangeably by vendors and inconsistently by buyers. The result is that procurement decisions get made on the basis of marketing copy rather than on the…
Read post →
The most common question I have been asked since Gravity went public is some version of: why are you not raising? After three shutdowns and a clean fourth bet, the assumption is that VC capital is the next move. The…
Read post →
Vibe AI was an AI friend product. From mid-2025 to early 2026, real users showed up, used it daily, and told me, in their own words, that they liked it. It also lost money on every active user every month. The full…
Read post →
The hardest shutdown across all three startups was the one with revenue attached. Specifically, the $100K-annual-run-rate slice of Super AI that was working in 2025 even as the surrounding product was structurally…
Read post →
Week 1 of Gravity is in the books. This is the tactical retrospective, written in the format I plan to use weekly: what shipped, what tools settled in, what decisions got made, what problems hit, week-1 metrics, and…
Read post →
The 10x rule from Zero to One applies at the product level: the whole product needs to be 10x better than the closest alternative for users to switch (Volt Equity, Thiel 10x rule). It is well-known and widely cited.…
Read post →
The public version of the Super AI postmortem is on this site already (read it here). It explains the structural failure: an all-in-one AI router built in 2024, killed in March 2025, in a category that was getting…
Read post →
From October 2022 to October 2023, I built MindWave out of Pune. Mental health, private sharing, support, journaling, group therapy on the roadmap, free professional community next to it. The product worked in the…
Read post →
People ask what the day looks like, the day a founder decides to shut down. The truthful answer is: it does not look like a day. It looks like weeks of rationalisation followed by a small moment that finally tips the…
Read post →
After three shutdowns, idea selection is no longer a creative act. It is a structured one. The cost of starting the wrong thing is now measured in years, not in weeks of energy. The rubric below is what I now run…
Read post →
Aryan Agarwal, building Gravity, from Bangalore. The default assumption when a 3x founder builds an AI agent platform is San Francisco. The default assumption is wrong for this company; it is also probably wrong for…
Read post →
This is a timing post, not a thesis post. The thesis lives in describe outcome, not workflow,workflow editors are the wrong abstraction once intelligence has moved into the system. This post is the narrower question:…
Read post →
Vibe AI was an AI friend. From March 2025 to February 2026, I built it, ran it, watched it work in week one, watched it stop working by week three, and then watched the unit economics confirm what week three had…
Read post →
Three startups, three shutdowns, three different missing checks. After the third one I realised the same framework would have killed every bad bet before it cost a year. This is the framework. None of the three…
Read post →
Super AI ran from March 2024 to March 2025. Twelve months. The pitch fit on a sticker: describe what you want, we'll pick the model. An all-in-one AI app that auto-routed each prompt to the best foundation model,GPT…
Read post →
This is month 1 of Gravity, in public. Public means: I write the retro before I have all the answers, name what shipped and what didn't, and let the kill thresholds I set at month start judge themselves. The point is…
Read post →