{"schemaVersion":"1.0","updatedAt":"2026-10-07","name":"Suryateja Kommuri","headline":"AI & Full-Stack Engineer","location":"London, United Kingdom","introduction":"I’m Suryateja Kommuri, an AI & Full-Stack Engineer in London. I build web applications, backend workflows and practical AI tools.","summary":"I built Voice AI Prospect Radar to help voice-AI vendors discover and review businesses that might benefit from automation. My work also includes website delivery and backend workflows for Haus of Estate. Alongside these projects, I’m exploring document retrieval; that work is still awaiting evaluation and conclusions.","url":"https://suryatejakommuri.com","opportunities":{"status":"Open to AI and full-stack opportunities in London or remotely.","roles":["AI Engineer","AI Solutions Engineer","Full-Stack Engineer"],"locationPreference":"London or remote","contactPrompt":"Send the role, company, working arrangement and a link to the brief.","emailSubject":"Engineering opportunity"},"capabilities":[{"title":"AI & retrieval","skills":["Document-grounded answers","Vector retrieval","Embeddings","Retrieval evaluation","Source citations"]},{"title":"Applications","skills":["TypeScript","Python","Next.js","React","Node.js","REST APIs"]},{"title":"Data & workflows","skills":["PostgreSQL","Prisma","Sanity","Content ingestion","Delivery queues","Consent handling"]},{"title":"Delivery","skills":["Vitest","Playwright","Docker","Git","Release handoffs","Technical documentation"]}],"selectedWork":[{"name":"Voice AI Prospect Radar","category":"Independent product engineering","contribution":"Business discovery, map-based review, explainable scoring, AI briefs and saved prospect tickets.","summary":"Find businesses, understand their fit and turn a shortlist into a review workflow.","outcome":"Built a complete discovery-to-review workflow: search businesses, inspect their scoring signals, generate a prospect brief and save a review ticket.","status":"Independent MVP","deliveryBoundary":"Previously deployed as a gated portfolio MVP. Current end-to-end availability has not been verified. The hand-designed scoring rubric has not been calibrated against sales outcomes; review signals are proxies. Commercial accounts and full CRM integration remain future work.","technologies":["Next.js","TypeScript","Google Places API","Prisma","PostgreSQL","OpenAI","Leaflet","Vitest"],"url":"https://suryatejakommuri.com/work/voice-ai-prospect-radar","updatedAt":"2026-10-07"},{"name":"Haus of Estate","category":"Product engineering","contribution":"Website performance, technical SEO, content validation and reusable page components.","summary":"Extending a property platform across website performance, content and discovery.","outcome":"Deployed technical SEO and homepage improvements, with reusable content tools and page components developed for ongoing platform work.","status":"Deployed work + ongoing development","deliveryBoundary":"Website improvements are deployed; further platform work remains in development. Client screenshots, unpublished listings and internal release details are omitted from this public overview.","technologies":["Next.js","TypeScript","Sanity","React","Playwright"],"url":"https://suryatejakommuri.com/work/haus-of-estate","updatedAt":"2026-10-07"},{"name":"Haus enquiry operations","category":"Backend systems","contribution":"Transactional persistence, consent handling, receipt validation and recovery tests.","summary":"Reliable enquiry capture, clear consent and resilient follow-up workflows.","outcome":"Implemented transactional persistence, receipt validation and independent delivery queues, with local tests for failure and retry behaviour.","status":"In development","deliveryBoundary":"This is implemented development work. Production delivery remains unverified. Client records, service configuration and internal rollout plans are kept private.","technologies":["TypeScript","PostgreSQL","Prisma","Vitest"],"url":"https://suryatejakommuri.com/work/haus-lead-operations","updatedAt":"2026-10-07"}],"explorations":[{"name":"Hybrid RAG workbench","category":"Retrieval exploration","contribution":"Retrieval comparison, reranking, evidence compression and evaluation tooling.","summary":"An unfinished comparison of classic retrieval and a CLaRa-inspired hybrid approach.","outcome":"Implemented comparison and evaluation tooling alongside the existing document-grounded baseline. Benchmark results, conclusions and a completed research write-up are still outstanding.","status":"Exploratory · not yet evaluated","deliveryBoundary":"Local development only. The comparison results have not been tested, and no conclusion or measured retrieval improvement is claimed. A completed research write-up is still outstanding. This is CLaRa-inspired work, not a reproduction of its full architecture. Structured table and figure ingestion and the newer Azure provider remain planned.","technologies":["Next.js","TypeScript","OpenAI","Pinecone","Document ingestion"],"url":"https://suryatejakommuri.com/work/hybrid-rag","updatedAt":"2026-10-07"}],"earlierWork":[],"contact":{"email":"kommurisurya@gmail.com","emailUrl":"mailto:kommurisurya@gmail.com?subject=Engineering%20opportunity","linkedin":"https://www.linkedin.com/in/suryateja-kommuri/","github":"https://github.com/suryak02","currentCv":"mailto:kommurisurya@gmail.com?subject=Current%20CV%20request","whatToInclude":"Send the role, company, working arrangement and a link to the brief."}}