Post by Prakhar Shukla

Excited to share a bit about the work I've been doing on PDR — our internal 𝗕𝟮𝗕 𝘀𝗮𝗹𝗲𝘀 𝗶𝗻𝘁𝗲𝗹𝗹𝗶𝗴𝗲𝗻𝗰𝗲 𝗽𝗹𝗮𝘁𝗳𝗼𝗿𝗺, built on Node.js, Express, MongoDB Atlas, and Next.js, with AI-powered enrichment to keep prospect data fresh and actionable. Some highlights from recent work: 𝗜𝗖𝗣 𝗘𝗻𝗴𝗶𝗻𝗲 — Designed an Ideal Customer Profile matching system: a cascading filter pipeline covering industry taxonomy, regions/countries, tech stack categories, and buyer personas, all feeding into a weighted scoring engine that ranks prospects against a company-defined ideal customer benchmark. 𝗦𝗰𝗼𝗿𝗶𝗻𝗴 & 𝗗𝗮𝘁𝗮 𝗜𝗻𝘁𝗲𝗴𝗿𝗶𝘁𝘆 — Identified and resolved several issues across the lead-scoring pipeline, including schema mismatches, enum inconsistencies between imported data and database models, and inconsistent scoring logic. Centralized everything into one auditable system where every score can be traced step by step. 𝗗𝗮𝘁𝗮 𝗤𝘂𝗮𝗹𝗶𝘁𝘆 𝗟𝗮𝘆𝗲𝗿 — Rebuilt the Excel/CSV import pipeline to handle messy real-world datasets at scale (100k+ rows). Invalid values are now flagged instead of rejecting entire rows, with missing-field diagnostics surfaced for review. 𝗔𝗜-𝗣𝗼𝘄𝗲𝗿𝗲𝗱 𝗘𝗻𝗿𝗶𝗰𝗵𝗺𝗲𝗻𝘁 — Improved the automated enrichment pipeline that detects tech stack and buyer intent signals for each prospect, making the data more reliable for sales teams. 𝗗𝗲𝗱𝘂𝗽𝗹𝗶𝗰𝗮𝘁𝗶𝗼𝗻 — Implemented multi-field duplicate detection (email, phone, name + account) to keep the prospect database clean as it scales. Grateful to my mentor Pavan Sengar sir for the guidance, and to our senior developers for the support and code reviews along the way. Learning a lot about what it takes to build systems that are explainable, scalable, and maintainable. Beno Support Technologies #SoftwareEngineering #B2BSaaS #BenoSupport #NodeJS #MongoDB #BuildInPublic