Company Description Vasco Tech is an AI-driven technology company focused on building transformative solutions across industries. Our mission is to harness the power of artificial intelligence to redefine how people learn, work, and innovate. With a vision rooted in progress and impact, we are creating tools and platforms that push the boundaries of what’s possible. We’re not just building technology - we’re shaping the future. Role Description We’re hiring a Backend AI Engineer to design and build the backbone of intelligent systems — from LLM orchestration to inference pipelines and distributed APIs. You’ll architect the backend infrastructure, implement reasoning workflows, and optimize for scale, speed, and reliability. You’ll work closely with the founding team to develop next-generation AI systems that think, adapt, and self-optimize. Key Responsibilities Architect and develop high-performance backend systems for AI orchestration, data retrieval, and reasoning. Build and optimize LLM inference pipelines and multi-model integration layers . Implement retrieval-augmented generation (RAG) , context management , and memory systems . Design APIs and microservices (REST / GraphQL) for reasoning and model coordination. Manage data storage and vector databases (Qdrant / Pinecone / Postgres / Neon). Build asynchronous, event-driven pipelines to handle large-scale requests and model calls. Deploy and maintain scalable infra using AWS, Docker, and CI/CD pipelines. Implement observability tools , evaluation metrics , and reasoning trace systems for debugging and improvement. Continuously benchmark and integrate new open-source or proprietary models . Must-Have Skills Strong in Python (FastAPI, asyncio, multiprocessing, or similar frameworks). Proficient in Node.js / TypeScript for service integration. Deep understanding of API design , scalable microservices , and async pipelines . Hands-on experience with vector databases (Qdrant, Pinecone, Weaviate) and Postgres / Neon . Practical understanding of LLM APIs (OpenAI, Anthropic, Mistral, etc.) and RAG architectures . Strong fundamentals in data structures , caching , concurrency , and distributed systems . Experienced with Docker , AWS / Vercel , and CI/CD automation. Ability to debug complex backend systems across data, infra, and AI layers. Preferred Skills Knowledge of LangChain , LlamaIndex , or custom orchestration logic. Experience with agentic workflows , model routing , or self-evaluating pipelines . Familiarity with model fine-tuning , embedding optimization , or vector search ranking . Exposure to observability stacks (Langfuse, Posthog, Sentry). Contributions to open-source AI infrastructure or published repos. Background in distributed system design or cloud-native architecture . Mindset & Culture Fit Builder-first — comfortable owning systems from architecture to deployment. Systems thinker — sees the interdependencies between models, data, and orchestration logic. Bias for speed and learning — iterates fast and optimizes later. Detail-obsessed — balances performance with correctness and traceability. Autonomous — thrives without constant direction, takes full ownership. Compensation & Growth Competitive salary. Opportunity to own core AI infrastructure and reasoning pipelines . Direct collaboration with the founding team on high-impact architecture decisions.
DO NOT APPLY IF YOU CAN'T BE BOTHERED TO READ THE ENTIRE JOB POSTING AND COMPLETE MENTIONED TASKS. Role: AI Generalist (Product Evaluation & Reasoning) Company: Vasco Tech Pvt Ltd Product: Allecta – a meta reasoning and orchestration layer that sits above AI models This is not a traditional job role. We’re opening a limited evaluation role for people who actively use AI and want to explore how advanced reasoning systems actually work in practice. If you use tools like ChatGPT, Claude, Gemini, etc. regularly – and feel that they often almost get things right but fail on deeper reasoning – this will be interesting to you. What you’ll do As an AI Generalist, you’ll: Explore Allecta hands-on across its core features Use it to solve real-world reasoning tasks (strategy, analysis, planning, problem-solving) Compare its outputs with other AI tools you already use Identify where the system feels meaningfully better – and where it breaks Be prepared to discuss your experience and insights during a short conversation This role is about thinking, judgment, and feedback – not coding, not ML theory, and not credentials. Who this is for You’ll likely find this role a fit if: You use AI tools daily for work or study You’re curious about how AI systems reason, not just what they answer You enjoy evaluating products and spotting patterns quickly You can articulate what worked, what didn’t, and why Backgrounds we expect to apply: Engineers Product managers Founders Consultants Analysts Students and researchers No specific degree, seniority, or technical background is required. Evaluation process (mandatory) This role is entirely evaluation-driven . Once you read this job post, you are expected to: Use the referral link provided in this posting to access Allecta ( https://www.allecta.ai/signup?ref=RTX9-ED2C ). Thoroughly explore the product across its core features Use Allecta for real reasoning tasks (not toy prompts) Actively look for: Errors Inconsistencies Confusing behavior Bugs or edge cases Your primary responsibility is to understand how Allecta works as a system and provide clear, structured feedback. Feedback requirement All feedback must be submitted through the in-product Feedback Form (Settings → Feedback) Feedback should be specific, actionable, and based on actual usage Generic or superficial feedback will not be considered Selection criteria Candidates who submit the most thoughtful and high-signal feedback will be shortlisted Shortlisted candidates will be contacted for further interview rounds Candidates who do not meaningfully use the product will be automatically disqualified During interviews, questions will be based directly on your experience using Allecta . Familiarity with the product is non-negotiable. This process is designed to reward curiosity, effort, and depth of understanding — not resumes. Outcome Some participants may be considered for deeper collaboration or future roles. Most will simply walk away with: Hands-on experience with a frontier reasoning system A clearer mental model of how multi-model AI orchestration works Early access insights most users don’t get How to apply Apply through LinkedIn Carefully read this description In your feeedback, briefly mention one specific capability of Allecta that stood out to you after exploring it. Applications that skip step 3 will not be considered.