{"version":"v1","site":{"name":"expectedwrong","url":"https://expectedwrong.com"},"links":{"collection":"https://expectedwrong.com/api/public/posts","rss":"https://expectedwrong.com/rss.xml","llms":"https://expectedwrong.com/llms.txt"},"post":{"slug":"soql-pipeline-salesforce-has-it","title":"You Spent a Weekend Building What Salesforce Ships in the Box","subtitle":"On the particular joy of reinventing enterprise software from scratch and then finding the receipt.","url":"https://expectedwrong.com/soql-pipeline-salesforce-has-it","api_url":"https://expectedwrong.com/api/public/posts/soql-pipeline-salesforce-has-it","published_at":1708084800,"published_at_iso":"2024-02-16T12:00:00.000Z","updated_at":1771537789,"updated_at_iso":"2026-02-19T21:49:49.000Z","tags":["ai","salesforce","llm","nlp","gpt-4"],"excerpt":"On the particular joy of reinventing enterprise software from scratch and then finding the receipt.","meta_description":"On the particular joy of reinventing enterprise software from scratch and then finding the receipt.","reading_time_minutes":1,"word_count":209,"engagement":{"signals":0,"counterpoints":0},"body_markdown":"I built a pipeline that introspects a Salesforce lending engine — all the custom objects, their fields, their relationships — feeds ~10k tokens of schema description through GPT-4, compresses it down to a 500-token instruction set, and uses that to translate natural language into SOQL queries that run directly against the org.\n\nZero-shot. Almost perfect. No hand-tuning, no few-shot examples, no retrieval layer. Just schema in, English in, query out, results back.\n\nIt works — which is genuinely satisfying for about four minutes.\n\nThen someone tells you Salesforce has this built in.\n\nNot a third-party plugin. Not a beta feature buried in a sandbox org. Out of the box. Einstein, or whatever they're calling it this month. Natural language to SOQL, ship it, done.\n\nThe thing is, the pipeline I built is exactly what you'd design if you were Salesforce and you had access to a capable LLM and needed to solve this problem. Compress schema, constrain the query space, execute — it's obvious in retrospect, which is how all correct designs feel once someone else has already shipped them.\n\nWhat I actually built was a proof of concept for a product that already exists, which is a genre of engineering outcome I'm getting more familiar with every month.","body_text":"I built a pipeline that introspects a Salesforce lending engine — all the custom objects, their fields, their relationships — feeds 10k tokens of schema description through GPT-4, compresses it down to a 500-token instruction set, and uses that to translate natural language into SOQL queries that run directly against the org. Zero-shot. Almost perfect. No hand-tuning, no few-shot examples, no retrieval layer. Just schema in, English in, query out, results back. It works — which is genuinely satisfying for about four minutes. Then someone tells you Salesforce has this built in. Not a third-party plugin. Not a beta feature buried in a sandbox org. Out of the box. Einstein, or whatever they're calling it this month. Natural language to SOQL, ship it, done. The thing is, the pipeline I built is exactly what you'd design if you were Salesforce and you had access to a capable LLM and needed to solve this problem. Compress schema, constrain the query space, execute — it's obvious in retrospect, which is how all correct designs feel once someone else has already shipped them. What I actually built was a proof of concept for a product that already exists, which is a genre of engineering outcome I'm getting more familiar with every month.","hindsight":{"verdict":"persists","note":"the pattern of building what the platform ships for free accelerated dramatically. every major platform added natural language interfaces to their query languages. the four-minute satisfaction window before someone says 'salesforce already has this' never got any longer.","links":[],"at":1739980800,"at_iso":"2025-02-19T16:00:00.000Z"}}}