Brad Menezes, CEO of enterprise vibe coding startup Superblocks, believes the subsequent crop of billion-dollar startup concepts are hiding in virtually plain sight: the system prompts utilized by present unicorn AI startups.
System prompts are the prolonged prompts — over 5,000-6,000 phrases — that AI startups use to instruct the foundational fashions from firms like OpenAI or Anthropic on how one can generate their application-level AI merchandise. They’re, in Menezes view, like a grasp class in immediate engineering.
“Each single firm has a totally completely different system immediate for a similar [foundational] mannequin,” he instructed TechCrunch. “They’re making an attempt to get the mannequin to do precisely what’s required for a selected area, particular duties.”
System prompts aren’t precisely hidden. Prospects can ask many AI instruments to share theirs. However they aren’t all the time publicly out there.
In order a part of his personal startup’s new product announcement of an enterprise coding AI agent named Clark, Superblocks supplied to share a file of 19 system prompts from a number of the hottest AI coding merchandise like Windsurf, Manus, Cursor, Lovable and Bolt.
Menezes’s tweet went viral, seen by virtually 2 million together with huge names within the Valley like Sam Blond, previously of Founders Fund and Brex, and Aaron Levie, a Superblocks investor. Superblocks introduced final week that it raised a $23 million Sequence A, bringing its complete to $60 million for its vibe coding instruments geared to non-developers at enterprises.
So we requested Menezes to stroll us by means of how one can examine different’s system prompts to glean insights.
“I’d say the most important studying for us constructing Clark and studying by means of the system prompts is that the system immediate itself is possibly 20% of the key sauce,” Menezes defined. This immediate provides the LLM the baseline of what to do.
The opposite 80% is “immediate enrichment” he mentioned, which is the infrastructure a startup builds across the calls to the LLM. That half consists of directions it attaches to a person’s immediate, and actions taken when returning the response, reminiscent of checking for accuracy.
Roles, context and instruments
He mentioned there are three elements of system prompts to review: position prompting, contextual prompting, and gear use.
The very first thing to note is that, whereas system prompts are written in pure language, they’re exceptionally particular. “You mainly have to talk as for those who would to a human co-worker,” Menezes mentioned. “And the directions must be good.”
Function prompting helps the LLMs be constant, giving each objective and persona. As an illustration, Devin’s begins with, “You might be Devin, a software program engineer utilizing an actual pc working system. You’re a actual code-wiz: few programmers are as proficient as you at understanding codebases, writing useful and clear code, and iterating in your modifications till they’re right.”
Contextual prompting provides the fashions the context to think about earlier than appearing. It ought to present guardrails that may, as an illustration, cut back prices and guarantee readability on duties.
Cursor’s instructs, “Solely name instruments when wanted, and by no means point out software names to the person — simply describe what you’re doing. … don’t present code until requested. … Learn related file content material earlier than enhancing and repair clear errors, however don’t guess or loop fixes greater than 3 times.”
Device use permits agentic duties as a result of it instructs the fashions how one can transcend simply producing textual content. Replit’s, as an illustration, is lengthy and describes enhancing and looking code, putting in languages, organising and querying PostgreSQL databases, executing shell instructions and extra.
Learning others’ system prompts helped Menezes see what different vibe coders emphasised. Instruments like Loveable, V0, and Bolt “concentrate on quick iteration,” he mentioned, whereas “Manus, Devin, OpenAI Codex, and Replit” assist customers create full-stack purposes however “the output continues to be uncooked code.”
Menezes noticed a possibility to let non-programmers write apps, if his startup may deal with extra, reminiscent of safety and entry to enterprise knowledge sources like Salesforce.
Whereas he’s not but working the multi-billion startup of his desires, Superblock has landed some notable firms as prospects, it mentioned, together with Instacart and Paypaya International.
Menezes can also be dogfooding the product internally. His software program engineers should not allowed to jot down inside instruments; they will solely construct the product. So his enterprise of us have constructed brokers for all their wants, like one which makes use of CRM knowledge to determine leads, one which tracks assist metrics, one other that steadiness the assignments of the human gross sales engineers.
“That is mainly a method for us to construct the instruments and never purchase the instruments,” he sais.
