PromptLayer

PromptLayer is looking for someone to help us teach the world about prompt engineering. If you love LLMs + building + want to share knowledge with the community reach out ✨ Responsibilities: - technical content writing - building llm projects Check out the PromptLayer blog to see some examples. Internship or part-time gig. Open for students and hackers 🍰

Starting a new tutorial serious of prompt engineering basics. Tip #1 is JINJA2. Extremely powerful. You might even remember it from the old web development days... Use it to swap out complicated Python code for simple jinja2 prompts → {% for item in user_info %} - {{ item.key }}: {{ item.value }} {% endfor %} The above will take in a JSON user_info and format it into a prompt. This let's you store more business logic in the CMS and less in code. Way easier to track. https://lnkd.in/eyDvRm6c

OpenAI just released fine-tuning for gpt-4o and gpt-4o-mini Try it out now on PromptLayer! Very powerful!

Everyone is talking about tool calling and structured outputs these days. It's the most important thing for a prompt engineer to know.. but every model does it a little differently. We put together a chart to compare the slight differences between models ✍️ Alternatively... you just store all your prompts in PromptLayer and they will be instantly runnable on any model. Even with tool calling 🔨 Read the full post https://lnkd.in/dv27M3-B

If you haven't seen a lot of killer LLM B2B apps or features in the wild, it's because LLMs are only the bottom of the iceberg when it comes to building a real-world product. LLM product development is similar to software 1.0 - maybe even slower because we don't have the proper tools yet. You still need great design, tests, infrastructure tooling, monitoring, alerting, etc. That's why it takes a long time to see the benefits. We have code copilots now, but things haven't improved 10x in how software engineering is done.

Very cool new release by @OpenAI. Structured outputs for the new gpt-4o model. This means you can now force JSON response in strict mode AND provide a schema for it to adhere to. No more "mis-using" tool calls to force schema. https://lnkd.in/eRjv5hUP

8 new Anthropic prompting tips I learned from Zack Witten (Anthropic's prompt engineer): Tip #1: Use XML tags to structure prompts Tip #2: Be specific with instructions rather than saying "be concise" Tip #3: Mimic desired tone of response in your prompt Tip #4: Proofreading your prompts matters Tip #5: Write prompts in a language you know well Tip #6: Use Claude prefills to force JSON response Tip #7: Use negative prompting sparingly Tip #8: Rely more on human messages than system messages Thanks to Anthropic's "Prompt Doctor" for the great workshop at AI Engineer 🙏 More details on each tip here in the full blog post: https://lnkd.in/edvcCWix Original workshop available here: https://lnkd.in/eVmAn9nc

You should be A/B testing your prompts! https://lnkd.in/eMiTXGt4

Recording is out from my talk at @arizeai Observe 🍰 I go through eval strategies such as e2e tests, unit tests, backtests Main points: 🎨 Prompt engineering is an art of tuning inputs 🥼 Domain experts are the key to evals 📝 Iterate, iterate, iterate https://lnkd.in/dMpmNR7g

*30 companies I’m an investor in are now hiring* I'd love to get you in front of these companies if you're exploring your next move. Reach out, and I will do my best to help. Full list here: 1) PromptLayer - prompt engineering platform (NYC) 2) Hero - AI selling (remote) 3) Natoma - non-human identity protection (Boston / Bangalore / Bay Area) 4) Pocus - signal based selling (Bay Area / NYC / remote) 5) Jitter - motion design (Paris) 6) Masterschool - network of career training schools (Tel Aviv / Germany) 7) Gelt - tax optimization and management (Tel Aviv / remote) 8) Luma - events platform (remote) 9) Gynger - tech payments on your terms (NYC / Tel Aviv) 10) SwagUp - swag platform (NYC) 11) Equals - next generation analysis and reporting (remote US) 12) Landa - fractional real estate investing (NYC / Tel Aviv) 13) Gestalt - human data interface (NYC) 14) ServiceBell - video live chat (remote) 15) Kick - self-driving bookkeeping (remote) 16) Yuzu Health - custom-built employer health plans (NYC) 17) Passionfroot - creator brand deals platform (Berlin / remote) 18) Deeptune - AI dubbing (NYC) 19) Alta Revenue Platform - revenue intel / business observability (Tel Aviv) 20) Trace Machina - automation infrastructure (Bay Area) 21) Modyfi - design platform (remote) 22) SellScale - outbound automation (Bay Area) 23) OwnID - password-free tech (Tel Aviv) 24) Gamma - AI-powered decks / sites (remote) 25) Balance - b2b transaction lifecycle (NYC / Tel Aviv) 26) Givebutter - fundraising platform (remote / Austin) 27) Hotplate - storefront for food drops (Bay Area) 28) Deel - global payroll (remote) 29) Daytona - dev environment manager (remote) 30) Haize Labs - AI safety (NYC) ✌️ follow me Ben Lang for more lists like these ___ #startups #careers #jobs