Raise the infrastructure for servicing models for the CV pipeline.

We are looking for a highly skilled AI application engineer and full-stack backend developer to build a production-ready AI-powered document validation, refinement, and approval workflow. This is not a simple prompt engineering role. We need someone who can design and implement a real AI application with strong backend architecture, Claude API integration, structured validation logic, audit trails, secure data handling, and human-in-the-loop review workflows. The system will act as an intelligent quality assurance layer for submitted reports and documents. It should review completed submissions, identify issues, improve content quality, apply business rules, protect sensitive information, and either approve the document automatically or route it for human review. The developer will be responsible for building a workflow that can: Pull completed documents, reports, or submissions from an external platform via API Analyze the full document, including structured answers, ratings, selections, narratives, comments, and free-text fields Perform semantic audits to detect logical conflicts, contradictions, missing information, vague statements, unsupported claims, or incomplete sections Validate that structured responses and written content are consistent with each other Apply custom validation rules, editorial guidelines, formatting standards, tone requirements, and business logic Detect, tokenize, mask, or securely handle PII, confidential data, and sensitive security-related information before AI processing where required Rewrite and enhance narratives, comments, and document sections for grammar, clarity, professionalism, consistency, and readability Preserve the original meaning, observations, and intent while improving the final output Standardize writing style across documents without making every report sound generic or over-normalized Flag content that appears inconsistent, fabricated, vague, incomplete, sensitive, or requiring human review Generate specific validation notes explaining why a document failed review and what needs to be corrected Automatically generate clarification or revision requests when more information is needed Support approval workflows where documents are: Automatically approved when confidence thresholds are met Routed to a human editor or validator for review Returned to the original submitter for revision or clarification Maintain a complete audit trail showing: Original submission Tokenized or masked sensitive data events AI findings and recommendations AI-rewritten content Human edits Approval or rejection decisions Final approved version Write approved and validated content back to the source platform through API integration The role also requires building an editor and final-decision workflow. Human reviewers should be able to inspect the AI’s findings, compare original and revised content, make edits, approve changes, reject recommendations, and finalize the document before it is sent downstream. Ideal experience includes: Strong Claude API / Anthropic API integration experience Experience building AI-powered document review, validation, editing, or compliance workflows Strong backend architecture skills Full-stack development ability Experience with API integrations, webhooks, queues, job processing, and database design Ability to design structured AI outputs, confidence scoring, rule-based validation, and human-in-the-loop review Experience with PII detection, tokenization, masking, encryption, access control, and secure AI data handling Experience building secure audit trails and approval systems Strong understanding of prompt design, but also the engineering skills to turn prompts into a reliable production system We are looking for someone who has already built serious AI applications, not someone who only writes prompts. The right person should be able to design the architecture, integrate with external APIs, manage document processing logic, protect sensitive data, build the review interface, and deliver a reliable workflow that can be used in production.

Development of firmware deployed on the gateway and management of direct interaction with PLC/equipment in the technical room (Modbus, BACnet, etc.).

We are looking for a 3D GenAI Engineer / AI 3D Pipeline Developer It is necessary to create a solution that can qualitatively generate 3D models from a single image or multiple images. It is important that this is not just a ready-made demo, but a clear and reproducible process: from the input image to a full-fledged 3D asset with mesh, geometry, textures, and the possibility of further use. What needs to be done: - test modern image-to-3D models and approaches; - determine which option is best suited for our task; - use Trellis, Hunyuan3D, or similar solutions; - if necessary, use Gaussian Splatting in the 3D pipeline; - configure the conversion of Gaussian Splat / splat representation into 3D mesh; - obtain usable geometry; - generate high-quality textures; - bring the result to a usable 3D asset state; - find the optimal balance between quality, generation speed, and pipeline complexity; - build a clear process that can be repeated for different images; - perform fine-tuning, LoRA, or other model adaptations for specific types of objects.

About the Project We are looking for an experienced AI Automation Engineer to design and build a secure, self-hosted AI platform that combines a local Large Language Model (LLM), Retrieval-Augmented Generation (RAG), and multiple AI agents to automate business workflows. This is a hands-on engineering role for someone who has experience building production AI systems—not simply integrating ChatGPT APIs. The goal is to create a private AI ecosystem capable of securely indexing company knowledge, answering questions using cited sources, processing meeting transcripts, and automating internal business processes. Responsibilities You will be responsible for: Designing and deploying a locally hosted LLM on a VPS or dedicated server Building a secure RAG pipeline using frameworks such as LlamaIndex or similar Creating document ingestion pipelines supporting PDF (including OCR), DOCX, TXT, XLSX and meeting transcripts Implementing document indexing, metadata management, deduplication, and versioning Developing AI agents for: Meeting transcription processing Automatic meeting summaries Action item extraction Client knowledge retrieval Building APIs or a simple web interface for querying the knowledge base Ensuring strict client data isolation and permission controls Implementing source-cited responses to minimize hallucinations Optimizing system performance, scalability, and reliability Writing documentation and deployment guides Performing testing and security validation Required Skills Strong Python development experience Experience with LLM frameworks RAG architecture experience LlamaIndex, LangChain, or equivalent Vector databases (Qdrant, Chroma, Pinecone, Weaviate, FAISS, etc.) Local/open-source LLM deployment (Llama, Mistral, Gemma, DeepSeek, etc.) API development (FastAPI preferred) Docker Linux server administration VPS deployment Git Authentication and access control Experience with OCR pipelines Experience working with structured and unstructured documents Fluent English What We’re Looking For The ideal candidate: Has built production AI systems from the ground up Understands RAG best practices Can work independently Thinks like a software architect—not only a developer Writes clean, maintainable code Communicates clearly Can recommend the best technologies instead of simply following instructions Project Type Freelance / Contract Remote Milestone-based Long-term opportunity for future AI automation projects Please Include With Your Application Portfolio of similar AI/RAG projects Examples of local LLM or AI agent implementations Estimated timeline Estimated project cost Hourly or fixed-rate pricing

There is a Telegram bot on aiogram/FastAPI (CRM for an event project) and a separate ManyChat bot in Instagram Direct for communication with clients.Task 1 — fix the logic of the Instagram Direct bot.The current bot is made according to a strict script: it works step by step (greeting → about the event → response to "too expensive" → response to "I'll think about it"), with 3 random text options for each step. The problem is that the bot does not understand the meaning of the message, but guesses the step number and sends a template that is off-topic. Because of this: when asked directly about the price, the bot does not provide numbers but sends general text the language jumps — sometimes Ukrainian, sometimes Russian essentially, it is a randomizer of templates, not a dialogueIn a week, the bot lost several live leads. I can provide screenshots of conversations as examples.Need: for the bot to actually analyze the content of the message (question about price, objection "too expensive", doubts "I'll think about it", non-standard question, etc.) and respond appropriately — based on AI (for example, Claude API), not according to a strict script. The tone should be lively, not robotic, and the language should be Ukrainian. Basic scenarios (price, discounts, handling "too expensive"/"I'll think about it") need to be preserved, but the AI should choose the appropriate response based on the context.Task 2 (optional, but also needed) — auto-posting in Instagram.Posting through the official Meta Graph API: reels, stories, posts, carousels. Tags in the bot: [reels] / [stories] / [post] / [carousel] + [geo] / [poll] / [countdown]. Multi-user panel /admin — adding/removing people, roles (posting / viewing / all), activity log. Dynamic price from Google Sheets — the bot pulls the current price from the "Price" column.Questions for the performer: Is it better to implement Task 1 (AI logic for Instagram Direct) on top of the current ManyChat via webhook + external AI processor, or as a separate bot on an architecture like aiogram/FastAPI, connected to Instagram through the official Meta API? Are you taking on both tasks together (Direct logic + auto-posting)? What is the total cost and timeframe for the entire scope (task 1 + task 2)?Requirements: Experience with Meta Graph API (Instagram) Experience integrating AI models (Claude API / OpenAI API) into chatbots Preferably — experience with aiogram/FastAPI or ManyChat webhooks Portfolio with similar projects