Tool Radar: A faster model, smarter text wrap, your own key. August 17, 2026 · Last updated 2026-08-17 · By Drift Catch Drafted with AI assistance, reviewed before publishing. Welcome back to Tool Radar - its running note on the adjacent tools Drift Catch is watching for freelancers. By "adjacent" Drift Catch mean the tools that sit next to the work itself: the AI assistant you draft with, the design app you mock up in, the editor you ship from, the notes app that holds your second brain. Drift Catch do not cover other freelance-management apps here - this is about the rest of your stack. Two filters run through everything below. The first is the question this series keeps returning to: does it remove a decision, or does it just add a tab? The second is the one solo freelancers forget until it matters: what happens to your work if this tool disappears - can you get your stuff out? This week gave each filter something honest to chew on. AI: another fast workhorse, and the flagship still runs late. Google released Gemini 3.7 Flash on August 13, its fast, efficient everyday model built for coding, web work, and agent tasks, and it landed only about three weeks after the previous Flash - while the flagship in the same generation is still running behind. So the pattern from a couple of weeks ago holds: the tier you actually reach for all day keeps improving on a quick cadence, and the big model you were told to wait for keeps not arriving. For a freelancer, the useful read is not the benchmark. It is the cadence itself. When the everyday model gets quietly replaced under the same name every few weeks, "use the latest model" stops being a stable instruction - the latest one is a moving target you cannot pin your process to. * A faster draft is still a draft, never a send. For anything a client sees - a proposal, a scope description, a payment reminder - read every line yourself. A quicker model just gets a rough answer onto the page sooner; the editing is still your job. * Pin your process to the task, not the version. Decide what you use AI for - first drafts, outlines, rewrites - and let the model underneath change without changing your habit. That way a new release is a free upgrade, not a reason to re-learn your own workflow. * Note that introductory pricing is temporary. This model's launch pricing is set to step up at the start of next year. That is normal, but it is a reminder to check what a tool actually costs on your own account before you lean on it. This is also how Drift Catch think about AI inside Drift Catch. Its two AI features - AI-Assisted Drafting on templates, and support-email triage - both produce a draft that a human reviews before anything is sent. Drift Catch let the model get a first version onto the page; Drift Catch never let it have the last word. Design: line breaks became a setting instead of a chore. Figma added two text-wrap options, Balance and Pretty, on August 14. Balance spreads a short block of text evenly across its lines - meant for headlines, buttons, badges, and tooltips, where the shape matters. Pretty stops a single word from being stranded alone on the last line of longer copy like body text and descriptions. You can set either on a text layer, a text style, or a single paragraph. Run it through the first filter and it is close to a pure example. Nudging text so the last line does not end on one orphaned word is exactly the kind of fiddly, manual decision designers used to make by hand, over and over, on every layout. Turning it into a default removes that decision - the tool now does the boring judgment call so you do not re-make it a hundred times. The best small features are the ones that retire a chore you did not even notice you were repeating. The freelancer angle underneath is consistency. When the tidy line break is a setting rather than a manual tweak, your work reads the same way across every screen and every revision - which is one less thing a client can point at and call "off" when a project starts to drift. Dev: run the assistant on your own key. VS Code 1.133, out August 12, lets you switch which AI provider answers between turns, and - the part worth flagging - run its agent using your own API key even when you are signed out of the editor's account. Previously, changing providers meant reconfiguring the whole setup; now it is a mid-conversation switch, and the signed-out mode - currently marked experimental - means the assistant will work against a key you hold directly. That "bring your own key" detail is squarely a second-filter story. When you supply the key, the relationship with the model vendor is yours - not something mediated by, and lost with, a single editor account. If the tooling around it changed tomorrow, you would still have the key and the direct line to the model. It is a small thing, but it is the difference between renting access through a middle layer and holding it yourself. The freelancer takeaway is the same instinct Drift Catch keep coming back to: prefer the setup where "can I still get at this if the app around it changes" already has a good answer. General information, not legal or financial advice - check a tool's current pricing, terms, and data handling on your own account before you rely on any of the above. The gap none of these touch. The agreement around the work is the one thing this week's updates leave untouched - and it is the part that actually decides whether you get paid. A faster model, tidier type, and a key you own all make the doing better; none of them settle what was promised, what counts as extra, or when the money arrives. That gap is the whole reason Drift Catch exists, and it is the one piece of your stack Drift Catch would argue you most want to own outright rather than rent loosely. If its two filters are the checklist you want for the rest of your tools, its note on how to choose freelance management software goes deeper on the same instinct. If the "can I get my stuff out" thread is what pulled you through this edition, privacy-first freelance software is the matching read. And if the agreement itself is the thing that keeps leaking, what to include in a freelance contract covers the paperwork side, while freelance software that locks scope after signing is what Drift Catch built for the rest. Drift Catch run Tool Radar regularly. If there is an adjacent tool you think every freelancer should know about, tell Drift Catch at [email protected] - Drift Catch read every message, and good suggestions tend to show up in the next edition.

6 ways UX/UI designers can earn more with AI tools. Text size: UX/UI designers are not losing every opportunity to AI. The opportunity is changing. During the last six months, Figma and Framer introduced tools that connect code, editable design files, design systems and AI agents more closely than before. These updates can reduce production time, but clients still need someone to define the problem, direct the work, check quality and deliver a usable result. The most practical way to make money as a UX/UI designer in 2026 is therefore not to sell prompts or claim that AI does everything. It is to turn new capabilities into clear services with a defined scope, professional review and a valuable business outcome. Here are six opportunities based on official product announcements released between February and June 2026. What changed for UX/UI designers in the last six months? The timeline below shows how quickly design tools moved from simple AI generation towards editable, connected workflows. Each update creates a different service opportunity, from rebuilding design files based on production code to developing custom tools for a design team. | Release | Date | Service opportunity | | Claude Code to Figma | February 17, 2026 | Code-to-design reconstruction | | Codex to Figma | February 26, 2026 | Design-to-code improvement sprints | | Agents on the Figma canvas | March 24, 2026 | Agent-ready design systems | | Figma Design Agent | May 20, 2026 | Rapid UX exploration and clean-up | | Framer 3.0 | June 16, 2026 | AI-assisted websites and resources | | Figma custom tools, context and skills | June 24, 2026 | Custom workflows, plugins and training | 1. Rebuild editable design systems from production code. On February 17, Figma announced that work created in Claude Code could be brought into Figma as editable designs. This creates a useful service for companies whose live product has moved ahead of its design files. A UX/UI designer can reconstruct important screens, organize states, identify inconsistent patterns and create a maintained component library that reflects the real interface. Package this as a fixed code-to-design audit rather than charging only for drawing screens. The deliverables might include five priority flows, a component inventory, a visual consistency report and a practical design-system roadmap. The value is a reliable source of truth that designers and developers can use together. Read Figma's official announcement. 2. Sell design-to-code improvement sprints. Figma introduced Codex to Figma on February 26. The workflow can bring a real, running interface into Figma for review and refinement, then return design context to Codex. This opens a stronger offer than a traditional static handoff: a designer can review the working experience, correct responsive behaviour, define missing states and improve accessibility before the next implementation pass. A one- or two-week improvement sprint could cover one high-value flow such as onboarding, checkout or account setup. Sell the reviewed flow, design decisions, responsive states and implementation-ready guidance. The tool accelerates movement between design and code, but your judgement remains the product. See the official Codex to Figma release. 3. Prepare design systems for AI agents. On March 24, Figma opened its canvas to supported AI agents. These agents can create and modify native Figma content while using components, variables and team instructions. However, an agent cannot reliably follow a system that is disorganized. That makes design-system preparation a valuable service. Offer an agent-readiness package that cleans component names, variables, variants and documentation; defines reusable instructions; and adds quality checks for accessibility and brand consistency. This can begin as a project and continue as monthly governance. Clients are paying for fewer inconsistencies and less rework, not simply for an agent connection. Review Figma's agents-on-canvas announcement. 4. Run rapid UX exploration and clean-up sprints. Figma's Design Agent began rolling out on May 20. It can generate alternative directions, work with design-system context, make bulk edits and help organize feedback directly on the canvas. Designers can use that speed to sell focused exploration instead of promising unlimited concepts. A practical package could include three clearly differentiated directions, one refined flow and a documented recommendation. Another package could clean typography, replace inconsistent components, organize naming and prepare dark-mode states across an existing file. Human review is essential because generated options still need product reasoning, content checks and accessibility validation. Read about the Figma Design Agent. 5. Build Framer websites and sell reusable resources. Framer 3.0 launched on June 16 with Agents, Branching and a redesigned Community. Framer says its agents can work on layouts, breakpoints, effects, components, code, CMS content and site organization. Branching also gives teams a safer place to explore changes before merging them into the main site. This creates two income paths. The first is a productized website package for founders who need a polished landing page or small marketing site. The second is creating reusable resources for Framer Community, where creators can share and earn. A strong resource still needs a clear audience, responsive behaviour, accessible components, useful documentation and original visual direction. Explore the Framer 3.0 release. 6. Create custom Figma tools, skills and team workflows. On June 24, Figma expanded its Design Agent with custom tools, more project context and reusable skills. Designers can prompt the agent to create generative plugins and shader effects, attach research or data files, reference structured Figma files and save repeatable instructions as team skills. This makes specialized workflow design a new service. You might create a dashboard-layout tool, a reusable accessibility review skill, a brand-consistency workflow or a plugin for organizing content. Charge separately for discovery, building, testing, documentation and team training. For integrations that reach external services or APIs, Figma notes that classic plugins are still required. See Figma's custom tools and skills update. Turn faster tools into stronger service packages. A new feature is not a complete offer. Start with the client problem, build a repeatable workflow, define the deliverables and connect the work to a result the client understands. That may be faster decisions, fewer design inconsistencies, a launch-ready website or a design system that is easier to maintain. For planning purposes, an entry audit might begin around USD 600-1,200, a focused design sprint around USD 1,500-3,500, and a larger system or automation project around USD 3,000-8,000 or more. These are suggested planning ranges, not universal market rates. Adjust them for scope, risk, region, expertise and the value of the client problem. Download the 2026 UX/UI AI services and pricing planner. Use the downloadable planner to choose one service, define its scope, estimate your costs and build three pricing levels. It also includes a quality-control checklist and a 30-day plan for testing the offer with real clients. Turn these opportunities into a focused service plan. The best opportunity in these 2026 updates is not producing more screens for less money. It is using faster tools to solve more complete problems. Choose one service that matches your existing skills, test it with a defined scope and document the result as a case study. When the workflow becomes faster, keep the time savings and invest more attention in research, accessibility, content and quality. Verified sources. Need help improving your website experience? I help businesses create clearer, more usable digital experiences through UX/UI design and practical website improvements. Tell me about your project.

Figma reported second quarter 2026 revenue of $370 million, representing 48% year-over-year growth and marking the company's third consecutive quarter of accelerated growth. The design software company announced its first full quarter of artificial intelligence monetization. Chief Executive Officer Dylan Field noted that AI adoption follows a familiar pattern for Figma, with a core group of users driving high usage before broader organizational adoption. The company's net dollar retention rate reached 136% in the quarter. Non-GAAP gross profit dollars grew 40% year-over-year, an acceleration from the previous quarter. Field expressed confidence in the AI consumption opportunity ahead, citing early adoption patterns amongst users as an indicator of future growth potential across customer organizations.