Notes on getting work done by agents
Comparisons, prices as of this month, and what actually works when you hand a task to an AI agent. Written by the people who build and test them.
AI Agents for Digital Marketing Freelancers
Solo digital marketing freelancers carry an unusually wide operational load: they do the client work, generate the reports, write the proposals, chase the invoices, manage the content calendar, and keep tabs on three…
AI Agents for Copywriters: Research, Briefs, Drafts
AI agents do not write your copy for you. They handle everything that has to happen before you write and after you submit, so you can spend more of your day doing the work that only you can do. Research, intake,…
AI Agents for Amazon Sellers: Listings, Inventory, Ops
Running an Amazon business means managing a constant stream of operational tasks: listings that drift out of rank, inventory levels that creep toward zero, buyer messages that need a reply within 24 hours, and Seller…
AI Agent Protocol Wars: MCP vs A2A vs ACP in 2026
Three open protocols are competing to become the connective tissue of agentic AI systems: MCP, A2A, and ACP. Each solves a different interoperability problem, and understanding the distinction determines whether you…
Gravity vs LangSmith: Platform vs Observability Tool in 2026
The short answer: LangSmith is LangChain's developer observability, tracing, and evaluation platform for LLM applications that engineers build and maintain in code. Gravity is an AI agent platform where…
OpenAI vs Anthropic: Agent Strategy 2026
OpenAI and Anthropic are the two labs most cited when people argue about where AI agents are heading. They share a lot. Both build frontier models, both let those models call tools, and both have moved toward agents…
How to Measure AI Agent Accuracy
Most people who ask how accurate their AI agent is are really asking the wrong question. They reach for a model benchmark number, a leaderboard score, a percentage from a vendor deck. None of that tells you whether…
AI Agent Regression Testing: A Practical Guide
Your agent worked yesterday. Then someone tightened a prompt, the model provider shipped an update, or a tool changed its output format, and now it quietly fails one in five of the cases it used to handle. Nobody saw…
AI Agent Quality Scoring Methods
You can score AI agent output quality four main ways, and most teams end up using a blend. Rubric-based scoring rates each output against named dimensions on a fixed scale. LLM-as-judge uses a strong model to apply…
AI Agent Pricing Trends 2026
The big shift in AI agent pricing for 2026 is this: the per-seat subscription, the model that built modern SaaS, is losing its grip. When software was a tool a person operated, charging per login made sense. Agents…
AI Agent for Sales Call Follow-up
You just got off a good sales call. The prospect leaned in, raised one real objection, and agreed to a next step. Then the next call starts, the day fills up, and by evening the follow-up is a vague intention instead…
AI Agent for Client Check-in Automation
Recurring client check-ins are the work that keeps relationships warm and accounts from quietly slipping away. The problem is rarely the message itself. It is the prep behind it: pulling up what the client has been…
AI Agent Evaluation Framework (Step by Step)
Most teams decide an agent is "ready" by trying it a few times and getting a good feeling. That is how a demo passes and production fails. An evaluation framework replaces the gut call with a repeatable process: you…
AI Agent Adoption Statistics (Updated Q3 2026)
If you want citable numbers on AI agent adoption, the honest starting point is this: the curve is steep, but the figures depend heavily on how each survey defines an "agent." The baseline is now near-universal.…
AI Agent A/B Testing Strategies That Don't Break Production
You changed the prompt, swapped the model, or added a tool, and now you want to know if the new agent is actually better. The honest answer is that you cannot know from a few hand-checked examples. Agent behavior…
AI Agent for Recurring Meeting Summaries
The weekly team meeting, the biweekly project sync, the monthly leadership review: these standing meetings are where a lot of work is steered, and where the record is usually worst. Someone takes notes one week,…
AI Agent for Payroll Prep
Payroll is the one process where a small input error becomes a personal problem for someone on your team. A missed timesheet, a new hire who was not added, a raise that did not carry over, and a paycheck is wrong,…
AI Agent for Month-End Close Checklist
The month-end close is rarely hard accounting. It is mostly coordination: forty small tasks across a few people, each waiting on a document, a reconciliation, or someone else finishing their part. The work that slows…
AI Agent for Meeting Notes to CRM
Everyone agrees the CRM should be up to date, and almost nobody keeps it that way. After a good call you mean to log the next steps, bump the deal stage, and note that the buyer mentioned a budget freeze until Q3,…
AI Agent for Journal Entry Automation
Every close has a stack of journal entries that look the same month after month. The rent prepayment gets amortized, the unpaid wages get accrued, the new laptop starts depreciating, and a handful of costs get…
AI Agent for Financial Report Generation
The monthly reporting pack eats a day that could be spent thinking about the business. Someone exports the trial balance, rebuilds the same profit-and-loss summary, updates the cash slide, recalculates the variance…
AI Agent for Expense Report Approval
Approving expense reports is a manager's least favourite recurring task. Most submissions are fine, so the work is mostly opening each one, scanning for a missing receipt or an obvious breach, and clicking approve,…
AI Agent for Action Item Tracking
Action items are where good meetings go to die. The discussion is sharp, the decisions are made, three people leave with clear tasks, and within a week two of those tasks have evaporated because nobody held the list.…
AI Agent for Accounts Receivable Follow-Up
Money owed to you does not collect itself. Invoices go out, due dates pass, and the polite reminders that would bring the cash in get pushed down the list behind work that feels more urgent. Weeks later you have a…
AI Agent for Accounts Payable Automation
Accounts payable is a queue that never empties. Bills arrive by email, by portal, and as attachments buried in threads. Someone has to read each one, key in the vendor and amount, check it against what was ordered,…
What Our First 500 Posts Taught Us
A while back I wrote a short reflection after a few hundred posts. We are now somewhere around 500. That number still feels strange to type, because almost nobody has read most of them yet. We are pre-launch, the…
The Content Engine Behind Gravity
This blog publishes a lot. If you have read more than a couple of posts here, you have probably wondered how a pre-launch startup ships this much without the writing turning to mush. The honest answer is that there…
Monthly Retro: Building Gravity in June 2026
I am Aryan, and I am building Gravity from Bangalore. This is the June 2026 retro, written the way I want to remember the month: honestly, in the open, before the polish sets in. Gravity is still pre-launch. There is…
Gravity Builder Program: Early Learnings
I want to be honest about where we are. Gravity is in private alpha, and the way we work with builders is early and still moving. So this is not a victory lap. It is a founder reflection on what we have learned…
AI Agent for Travel Booking Optimization: A Workflow
Booking a trip well is a tab-refreshing chore. You compare a dozen fares, keep checking whether the price dropped, weigh a cheaper flight against a brutal layover, and then double-check it all fits the budget and, if…