Lokman Efe

Next.js, React Native, TypeScript

• Shipped end-to-end features for a production AI platform using Next.js and React Native, spanning web, mobile, APIs, persistence, and production rollout.
• Designed evaluation workflows for LLM routing and multi-agent coding harnesses, using controlled production-like runs and event-level telemetry to isolate model, orchestration, infrastructure, and measurement failures.
• Improved LLM routing and streaming reliability by parallelizing classification, eliminating redundant data loading, bounding model context, and strengthening failure handling under concurrent workloads.

Next.js, React Native, TypeScript

• Built and launched the web app from scratch in Next.js, owning frontend architecture, Figma-to-production implementation, and MVP release readiness.
• Maintained and stabilized the React Native mobile app through bug triage, fixes, and cross-platform flow alignment across ongoing feature work.
• Worked directly with backend, product, and management to shape MVP scope, resolve launch blockers, and align web/mobile user flows.

Next.js, Python, TypeScript

• Worked across Prymatica, 1Lookup, and 1Capture, shipping full-stack features in a fast-moving async environment.
• Built Stripe billing flows for 1Capture, including trial pages, pricing changes, revenue-based subscriptions, and production-facing UI/UX fixes.
• Built Python automation for Prymatica to generate outbound email templates and support lead-generation workflows.

Node.js, Python, GraphQL, AWS

• Integrated 30+ third-party platforms using OAuth2/Nango, handling auth, data sync, rate limits, pagination, schema design, and API documentation.
• Built multitenant RAG/data retrieval workflows with indexing, BM25/ranking logic, vector search flows, and evaluation pipelines for AI-assisted product features.
• Created OpenAPI documentation and a Stripe-like API key system to expose platform capabilities to external users.

React, TypeScript, Electron

• Helped build LingoEdit, an editor-based AI translation app that launched #2 on Product Hunt.
• Built new editor types including a voice-AI dictation editor, integrated multiple AI providers, and ported the web app to an Electron desktop app.

Python, OpenCV, YOLO, Linux

• Built computer-vision and OCR pipelines with Python, OpenCV, YOLO, Roboflow, Tesseract, PaddleOCR, EasyOCR, and Linux-based ML workflows.
• Processed 20,000+ labeled records and contributed to multimodal AI and edge-AI prototypes using camera input, sensors, Jetson Nano, and Arduino.