mayop100

Tasklet | Head of Design Engineering, Software Engineer (multiple), Customer Success Engineer | San Francisco, CA | ONSITE | Full-time

Tasklet is building the AI platform companies use to run their business on agents. You describe an outcome in plain English and our agents do the work across your SaaS tools, APIs, browser, files, and internal systems — immediately, or in the cloud 24x7 on a schedule or in response to events. Every agent gets its own cloud computer, filesystem, database, and browser. We're a team of 12 in SF, founded by Andrew Lee (Firebase founder) and Jonny Dimond, backed by $42M from USV, Lightspeed, YC, and others.

Head of Design Engineering ($300–450K base, 0.35–0.70%) — IC role owning our product experience end to end: decide how Tasklet should work, design our UX, and personally ship it in production. Looking for an experienced ex-founder builder with exceptional taste and real design & frontend chops.

Software Engineer — multiple roles ($220–400K base, 0.15–0.60%) — agent harness, context management, sandboxed execution, model serving, storage, Kubernetes, mobile & web apps. Deep systems fundamentals matter more to us than an AI résumé. Mid-level through Staff+.

Customer Success Engineer ($130–180K base, 0.05–0.15%) — own customer relationships & support our users. Help our customers use Tasklet more effectively and feed what you learn back into your core offering. We're looking for an AI power-user who's great with customers. ~1–5 years professional experience. No engineering background required.

All roles are in person at our office in SF -- no exceptions -- and we're not sponsoring visas at this time.

Full details here: https://tasklet.ai/careers — to apply, email me ([email protected]) with your resume and a video under five minutes that includes a demo of the most impressive thing you've built recently. See careers page for specifics.

Lets add another simple game! I vote flappy bird.

Add the current weather in my area to the top right, live-updating.

Tasklet | Staff Software Engineer - AI Agents | San Francisco, CA | ONSITE | $260k-$330k + 0.8-1.6% equity

At Tasklet (https://tasklet.ai/), we're building the world's most powerful AI agent platform for business automation.

We're looking for a staff-level software engineer to share technical leadership of our core agent infrastructure. You'll architect systems for multi-step agent orchestration, state management, trigger systems, and seamless integration with thousands of APIs.

Our stack: TypeScript, React, GCP, Kubernetes, Claude

Looking for: 5+ years experience building production-scale systems, proven technical leadership, experience building AI agents, exceptional programming skills across the full stack.

Compensation: $260k-$330k salary + 0.8-1.6% equity, competitive benefits including medical, dental, vision, 401k, 4 weeks PTO

Apply: https://shortwave.notion.site/Open-Positions-at-Tasklet-2a36...

This is the basic idea we built Tasklet.ai on. LLMs are great at problem solving but less great at cost and reliability — but they are great at writing code that is!

So we gave the Tasklet agent a filesystem, shell, code runtime, general purpose triggering system, etc so that it could build the automation system it needed.

If you’re looking for a (non-meta-owned) alternative, check out our startup Tasklet (tasklet.ai)

[founder here] Zapier, n8n, Agentkit, etc all have the user define a flowchart in software. This is a lot of work to set up and brittle -- can't work around edge cases & nuance.

We use an agent for execution -- so the LLM decides what to do at every step. No flowchart at all -- just text instructions.

Shortwave | Onsite in SF | Full-time | shortwave.com

We’re building Cursor for business communications (starting with email)

We’re currently growing revenue very quickly and are looking to more than double our team in the next 3 months. Hiring AI agent engineers, product engineers, frontend engineers, designers, and customer service & growth content roles. Aggressive comp.

I previously founded Firebase & sold it to Google. Shortwave is going to be much bigger.

Full details: https://shortwave.notion.site/Open-Positions-at-Shortwave-1a...

Shortwave | Onsite in SF | Full-time | shortwave.com

We’re building the [“strong form”](https://www.youtube.com/watch?v=2ZpJXHiPwtQ&t=498s) AI agent app for business communications. We’re focusing on email initially, as it’s (1) ubiquitous (2) open and (3) has a diverse set of valuable business data. In the future we plan to expand beyond email to include all essential business communications tools, including Slack, Teams, CRMs, document platforms, project management tools, and so on.

Most people’s experience with AI today involves copy & pasting content back and forth from ChatGPT. Not only is this a terrible UX — it severely limits the transformative potential of AI.

Our goal is to unlock 10x (or greater) productivity gains for the average business user by bringing cutting edge AI together with their business communications in a single app designed for effective human-AI collaboration. With Shortwave, users can give the AI access to all of their data and then allow it to take action while still maintaining total control over its output.

I previously founded Firebase and sold it to Google. This time we're taking it to IPO :)

Hiring for multiple engineering roles. See full listings here: https://shortwave.notion.site/Open-Positions-at-Shortwave-1a...

This makes it onto the front page of HN like every couple years.

Jacob (its creator) is now my cofounder at Shortwave (https://www.shortwave.com). Check it out if you want to see his latest work :)

Hi HN -- founder here -- I believe we have something more powerful here for email than anyone out there (far more capable than the Bard or Duet AI features Google recently released, for example).

I go into some detail about how the whole system works on Twitter here: https://twitter.com/startupandrew/status/1704564773831197032

Excited to hear what ya'll think!

I go

Right now we use an LLM to extract a textual description of your writing style from past emails, and then we use that in the prompt.

The nice thing about this is that it's editable, so the user can customize the style to be the style they want, not necessarily exactly the style they have.

We're also investigating doing per-user model training for more refined voice... not launched yet though (and comes with tradeoffs).

Happy to swap notes -- email me: [email protected]

Non-Gmail support is on our list!

There's definitely more we want to do here - including prompting you proactively in helpful ways. This is our v1 -- stay tuned for v2 :)

Bard's integration with Gmail is extremely basic -- it's just using regular Gmail search to include a few relevant emails in its prompt and then trying to answer your question. There's no AI on the email side.

We launched a much more capable Gmail + AI assistant this morning here: https://news.ycombinator.com/item?id=37585990#37586627

We're using embeddings + vector DB + x-encoder + GPT4 to deliver a much smarter & capable assistant.

Bard's integration with Gmail is extremely basic -- it's just using regular Gmail search to include a few relevant emails in its prompt and then trying to answer your question. There's no AI in the email side.

We launched a much more capable Gmail + AI assistant this morning here: https://news.ycombinator.com/item?id=37585990#37586627

We're using embeddings + vector DB + x-encoder + more to deliver a much smarter & capable assistant.

Boom! Glad you like it - and thank you for the kind words :)

[Co-founder of Shortwave here] I know a lot of folks are launching “AI Assistants” right now – but ours isn’t just a “chat with your PDF” thin shim on GPT4. We’ve got some serious infrastructure behind this.

Here are some notes on our architecture:

- We use LLMs at multiple places to choose what data to pull at each step. We use an additive approach rather than a chaining approach to avoid error propagation. We use GPT3.5-turbo with a bunch of hand-rolled prompts for most of this.

- We’re using InstructorXL + Pinecone running on GCP for vector-based search. We combine this with more traditional search methods backed by Postgres & Elasticsearch, to give the assistant the ability to fast searches of multiple types. We use a x-encoding model trained on open source Q&A data from Bing for scoring & reranking to allow us to combine multiple data sources and determine what makes the most sense to feed into the final prompt.

- We hand-rolled a bunch of rule-based algorithms and heuristics on top of the LLMs to deal with email-specific corner cases and other issues we couldn’t resolve reliably in prompts

- Our user-facing output is generated with GPT4.

This enables a bunch of capabilities that other AI assistants can’t match:

- Way better search — Ask a question and get a succinct direct answer, including finding emails that would be tough for you to find through traditional search (ie. you can’t remember a keyword to use).

- Scheduling – Since we can dynamically pull in multiple types of data, we can access calendar data at the right time to help you schedule meetings.

- Analyze across multiple emails & types of data - The assistant can synthesize answers across multiple emails, your calendar, setting, etc to give you an answer (eg. “What are the top 5 issues that customers emails support about last week”, “what are some meeting times that work for me and the other people on this thread”),

- Write in your voice — the assistant can automatically learn your style and tone based on your sent emails. This means it actually sounds like you and, while it still requires some tweaking occasionally, it’ll save you a lot of time.

- Summarize & translate – it can dynamically access the data you have on your screen right now if you reference it, so it can help you with whatever you’re reading.

A note on privacy: We take privacy very seriously. We’re running everything above on our own GPUs + using OpenAI for final outputs. We aren’t training any models on user data.

We’ve put a lot of thought and effort into this one – I hope you like it – either way, let me know what you think in the comments below!

-Andrew

Shortwave | Senior or Staff Software Engineer | ONSITE in San Francisco | $140k-$250k base + generous equity

I previously cofounded Firebase (YC S11) and sold it to Google, and most of our team is ex-Google / ex-Firebase.

Shortwave (https://www.shortwave.com/) is building a cybernetically enhanced email client. We're bringing the power of LLMs to your email to help you get things done. On top of this, we're bringing users a modern messaging UX and best-in-class productivity features. Email is an enormous market (4.5 BILLION users), and we're working to build the next unicorn productivity tool.

Imagine an AI virtual assistant with perfect knowledge of every message, attachment, receipt, newsletter, etc you've ever sent or received. Your AI assistant can answer questions about your email ("which candidates in my inbox have experience with Kotlin?") and act on your behalf ("Draft responses asking them to schedule a time with me").

We are building a world-class AI software team. We use a modern tech stack (Typescript, Kotlin, K8s, React, cutting-edge ML stack) and move ship 5x+ times per day.

We want the best and are willing to pay for it. We like generalists with strong CS fundamentals, app development experience, scrappiness, speed, and grit.

Come help us build the future. Email if interested: [email protected]