TasksMind

@ThangDo102 The Sf Vibe!!

new vlogggggg

A year ago, I was doing internships and getting paged for repetitive engineering issues. Now we’re in SF building @TasksMind , AI agents that investigate incidents and help engineers sleep. Crazy seeing @forumventures post about us. We’re just getting started 🚀

Team bonding matters way more than people think. Building in SF hits different when you’re: shipping together going to events together meeting other founders + engineers debating ideas at 1 AM living the startup grind in person Remote is great. But in-person energy compounds fast. SF really is a builder city.

@Vatsalpandya333 @CDebnath @ksyed1125 @ThangDo102 Let’s Go Team♾️

@ThangDo102 🚀🚀

San Francisco, I'm here. 🌁 I joined TasksMind as a co-founding engineer because I believe the future of on-call is autonomous, AI agents that take the alert, investigate, and ship the fix before a single engineer loses sleep. We're early. We're building fast. And now we're in the city where this kind of thing gets built. If you're working on infra, AI, or reliability tooling, let's connect. Always down for a coffee and a good technical conversation.

TasksMind is officially fully in SF full-time now. A few months ago this was just late night calls from Nebraska. Now it’s founders under one roof building AI agents for developers on call in the middle of San Francisco. Still feels surreal.

@Vatsalpandya333 👀

is this a better logo? @TasksMind

@Vatsalpandya333 🐐

Mumbai to San Francisco. 18-year-old me came to the U.S. with two suitcases and no roadmap. Now I’m in SF building an AI startup with my cofounders. Crazy how life works sometimes.

@Vatsalpandya333 @ksyed1125 @CDebnath 🐐s

Had to touch the grass with the team San Francisco were too good

@ksyed1125 🐐

I’m Kashish. Just moved to SF from Madhya Pradesh, India via Nebraska. I’m 22, the co-founder and CTO of TasksMind, and I ended up here because of an Instagram DM. A friend I met during orientation messaged me about an idea. Honestly, the idea was questionable. But I replied anyway. We eventually pivoted into what TasksMind is now, and somehow that one message brought me from Nebraska to SF. That DM changed my life. A few things about me: • I listen to music in five different languages on rotation. Bollywood, K-pop, C-pop, Thai pop, Western pop. Algorithm is confused. • I alter my own clothes. Just finished crocheting a sweater for my boyfriend. • I’m a CTO building AI agents, but my hot take: AI isn’t going to take over. Engineers won’t let it. We know how broken these systems actually are. If you’re in SF and want to grab coffee, talk on-call infra, agents, fashion, or yell about how overhyped AI doom is, say hi.

At 18, I left Mumbai and landed in Nebraska with two suitcases, a laptop, and no safety net. No network. No shortcuts. No famous last name. Just a student visa and a ridiculous belief that I could build something here. For years, I was trying to get picked. Picked for internships. Picked for interviews. Picked by companies. Picked by people who had no reason to bet on me yet. Then I stopped waiting. A few weeks ago, UNL’s School of Computing published a story about me, my cofounders Kashish Syed @ThangDo102 , and @TasksMind . Now Nebraska Today picked it up too. That one hit different. Because this is not just an article to me. It’s a receipt. A receipt for every late night. Every rejection. Every interview I thought I ruined. Every time I questioned if I belonged here. Every call home where I acted like everything was fine. TasksMind started from one thing I kept seeing across real engineering teams: on-call engineers are not burned out because every incident is impossible. They are burned out because too much of it is repetitive. Same alert. Same logs. Same investigation loop. Same fix. Again and again. So we’re building AI agents for developers on call. Agents that investigate incidents, pull logs, understand context, identify root causes, propose fixes, test them, and open clean PRs. Not dashboards. Not summaries. Not another AI wrapper. Execution. When a predictable alert fires at 2 AM, the engineer should not have to wake up just to repeat the same investigation for the fifth time that month. I came here trying to earn a seat at someone else’s table. Now we’re building our own. From Mumbai to Nebraska to San Francisco. From student visa to founder. From trying to get picked to building something people are starting to notice. Still early. Still hard. Still a lot to prove. But we’re not waiting anymore. We’re building. Nebraska Today story: news.unl.edu/article/pandya…

@Vatsalpandya333 LFG!!!

Came to Nebraska from Mumbai at 18. Now my university just published a story on me and the startup we’re building. TasksMind: AI agents for developers on call. Still feels weird typing that. Article: here Came to Nebraska from Mumbai at 18. Now my university just published a story on me and the startup we’re building. TasksMind: AI agents for developers on call. Still feels weird typing that. Article: here computing.unl.edu/news/pandya-pr…

Our AI agent just opened a PR to fix a production incident. No human touched it. API latency spiked. Alert fired. @TasksMind woke up before any engineer did. It read the alert. Traced the spike. Found the root cause. Drafted the fix. Opened the PR. Engineer time spent: zero. Human resolution time: 1-2 hours. This is not a summary. Not a dashboard. Not “here are the logs, good luck.” An actual agent. Doing the actual work. On-call is broken. Engineers are the most expensive incident response tool a company has. And they’re getting paged for things an agent can handle end-to-end. We built the engineer that never sleeps. This is what that looks like. If you’re an engineer who’s lost weekends to incidents or a CTO burning out your on-call rotation This is for you.

A year ago, I was balancing classes, work, and building late nights, constantly asking myself how far I could really go as an international student. Today, I’m proud to share that @TasksMind is now a VC-backed startup, backed by @forumventures . TasksMind started from a real pain. Being on call. 2 AM alerts. The same incidents. The same manual firefighting. We’re building AI agents for developers on call. When an alert fires, our agents investigate incidents end to end and take action. Not just summaries. Real execution. This milestone means more than capital. It’s a signal that we’re building something real. Grateful to our partners at Forum Ventures, especially NealSarraf, JamesMurphy, Toni Agbaje-Ojo, and Dave Coen for the support and belief. None of this happens without the team. Huge credit to my co-founders Kashish Syed, TylerFrisinger, and Thang Do for building relentlessly alongside me. As an international student, going all-in isn’t optional. There’s no Plan B. No safety net. If you’re an international student sitting on an idea and waiting, this is your sign. Back to building.

@Vatsalpandya333 🫶🫶

Lfggg Wish Us Luck Gang!

@Vatsalpandya333 🔥🔥

Last night, someone I know got woken up by a fire alarm. Not because there was a fire. Because someone burned toast. Whole building awake. Adrenaline. Chaos. For… nothing. That’s on-call engineering today. We’ve built systems so sensitive that humans are the circuit breaker. Every alert, real or fake, routes to a sleepy engineer at 2am. And the industry’s big idea? “Let’s give them an AI copilot.” No. That’s like giving the fire alarm a nicer ringtone. The problem isn’t that engineers need help thinking. The problem is they’re doing work that should not exist. On-call is: • Repeating the same investigation • Running the same playbooks • Restarting the same services • Writing the same postmortems Again. And again. And again. At 2am. If AI still needs a human to click every button, nothing changed. Real progress is when: • Alerts triage themselves • Incidents investigate themselves • Known fixes execute automatically • PRs open without asking permission Humans should be the exception, not the default. We didn’t give elevator operators better tools. We removed the job. On-call is heading the same way. Uncomfortable? Good. If you think AI should only “assist” engineers forever, explain why. If you’ve lived on-call, you already know where this ends. Where do you land?

On Friday, it finally happened. Standing on stage. Cameras everywhere. Our company on the screen behind me. Not an idea. Not a concept. A real product. A real company. @TasksMind AI agents that take over on call work for engineers. Alerts. Investigation. Fixes. Execution. This started with a simple belief. If work is repeatable, humans shouldn’t be waking up at 2 AM to do it. Months of building. Late nights. Hard pivots. Moments of doubt. And on Friday, we showed what we’re actually shipping. Grateful for the builders, mentors, and early believers who helped make this real. Still day one. Big shout out to @dedaluslabs X @theresidency @itsCathyDi @NickyHeC01 @palashaw_ Back to building.

Week two in SF changed everything. Came here from Nebraska/Mumbai with one belief. Agents should finish your work for you. No laptop rush. No late night pull requests. Just talk to your agent and it executes. @dedaluslabs @ycombinator S25 @theresidency proved the vision. @TasksMind feels stronger than ever. Right people. Right speed. Right momentum. This is only the start. Let’s build. #AI #YCombinator #Founders #TasksMind