# Sai Charan Jogu Portfolio of **Sai Charan Jogu** (publishes as Sai Charan). Independent Product Consultant at AlphaX Studio in Hyderabad, India. Canonical site: https://saicharan.design I go into an operation and find the step where a person waits, or repeats, or decides slowly. I put a cost on that step. Then I build the system that removes it, and I run the acquisition motion that fills it. Thesis on the homepage: the model was never the bottleneck; the human loop around it was. ## Headline metrics - **4**: Clients Sourced & Closed (B2B / government) - **5**: AI Systems Shipped 0->1 - **INR 1Cr+**: Ad Spend Managed - **2-3x**: Conversion or throughput lift measured per system - **50+** eval scenarios - **42** depth interviews on the flagship deployment - **10+** usability tests - **4-6** pod size - **2-week** sprint length Eval results from the Bookmore harness, reported on that case study: correct answers 64% -> 91%; incorrect information 41% -> 3%; correct tone of voice 23% -> 89%. ## Experience ### Independent Product Consultant, AlphaX Studio (2023-Present) I find the slow or expensive step in a client operation and build the system that replaces it, owning the work from the first conversation to production. ### Product Management Intern, HUEMN (2024) I redesigned the activation funnels, cutting time-to-value by 30%. ### Founder, GetCam (2020-2022) I built a camera rental marketplace and grew it from zero to more than 1,200 rentals. ## Education and recognition - **MBA, Product Management**, Kairos Business School (2024) - **B.Tech, Information Technology**, BVRIT Narsapur (2023) - **Kairos Fellowship**, Selected founder cohort (2023) - **Speaking:** IEEE SB OUCE Founders Talk. I gave a talk about startup strategy to engineering students. https://ieeesbouce.org/2024/10/12/founders-talk-by-sai-charan-jogu/ Toolkit: RAG pipelines, Agent orchestration, Eval-gated releases (LangSmith, DeepEval), Meta Ads, Apollo.io. The Bookmore eval harness also names Promptfoo. ## Deployments ### Bookmore (2024) Photographers lost most of their booking-intent leads to response latency. A reply that arrived after 15 minutes had already lost the client. I measured the loss, then built an agent orchestration system that qualifies the lead in Instagram direct messages in under 10 seconds. I proved it against a 50-scenario eval harness before any studio depended on it. - **Inefficiency:** 60% of booking-intent leads died in the reply queue - **System:** Agent orchestration that answers in under 10 seconds - **Response latency:** 15+ min -> under 10s - **Lead-to-book conversion:** 12% -> 38% - **Eval scenarios engineered:** 50+ - **Page:** https://saicharan.design/bookmore ### Fotos CRM (2024) Manual delivery workflows caused a 48-hour turnaround and limited each photographer to one order a day. I built an AI automation pipeline that cut delivery time to 19 hours and doubled order capacity. - **Inefficiency:** Manual delivery capped photographers at 1 order per day - **System:** AI automation pipeline for the delivery workflow - **Delivery time:** 48 hrs -> 19 hrs - **Order capacity:** 2x - **Retention:** +25% - **Page:** https://saicharan.design/fotos ### Reconciliation Agent (2025) Manual reconciliation was a slow and error-prone bottleneck in internal finance and operations. I deployed an actor-critic harness: a policy agent proposes the matches, and a critic model validates each one against the business rules before acceptance. This cut false-positive errors and the manual review load. - **Inefficiency:** Manual reconciliation was slow, and its errors were expensive - **System:** Actor-critic harness that validates before it accepts - **Architecture:** Actor-critic - **Harness:** OpenClaw, Hermes - **Page:** https://saicharan.design/reconciliation ### Skyvisa (2024) Six out of ten applicants dropped out of the visa form because validation errors arrived too late to act on. I built an AI-assisted form experience with predictive field validation and ran 10+ usability tests against it. - **Inefficiency:** 60% of applicants abandoned the form at validation errors - **System:** Predictive field validation built into the form UX - **Task completion (usability testing):** 40% -> 64% - **User satisfaction (n=12):** 4.2/5 - **Page:** https://saicharan.design/skyvisa ## Archive - **DRDO Defense System** (2024): NDA-restricted engagement diagnosing and rebuilding a workflow bottleneck in an air-gapped, security-audited environment. No public case study. - **GearHub** (2024): Asset and inventory control for production studios. Accountability chains and QR tracking. - **Magic Upload** (2024): Zero-loss ingestion pipeline for high-volume wedding photographers. - **VJOC** (2023): Indo-Japan academic centre. Headless WordPress and a Next.js frontend, 95+ Lighthouse. - **Prooduct** (2023): Fractional CPO service. A repeatable 8-week framework from idea to shipped MVP. - **GetCam** (2020): Camera rental marketplace. 1,200+ rentals, and a hard pivot from P2P to B2C. ## Revenue systems The pipeline is the same system as the product. ICP definition, Apollo.io outbound with A/B tested cadences, and Meta paid acquisition managed against booked calls all feed the qualification agent described under Bookmore. The clients were sourced and closed founder-led. - **INR 1Cr+**: Meta ad spend managed for clients - **4**: B2B / government clients sourced and closed - **under 10s**: Time from inbound lead to qualified reply ## Writing - **Total impressions:** 368K - **Unique reach:** 185K - **Follower growth:** +37% - [The Cola Wars: Reliance vs Coca-Cola](https://www.linkedin.com/feed/update/urn:li:activity:7264113845219016706): 180K impressions. Price-led market entry against an incumbent - [Rapido: The Localization Playbook](https://www.linkedin.com/feed/update/urn:li:activity:7262432765529194496): 47K impressions. Distribution built for a market, not ported into it - [Starlink in India: Market Disruption](https://www.linkedin.com/feed/update/urn:li:activity:7338765830790201344): 25K impressions. Regulatory friction as the real go-to-market constraint ## Contact - Email: saicharanjogu0@gmail.com - Calendly: https://calendly.com/saicharanjogu0 - LinkedIn: https://www.linkedin.com/in/saicharanjogu/ - GitHub: https://github.com/saicharanjogu/ - Resume: https://saicharan.design/SaiCharan_CV.pdf