Founding AI Engineer (Agentic AI)
6y relevant experience
Executive Summary
The candidate Chakrabarty is a strong candidate for the Founding AI Engineer role with 8 years of production AI experience, genuine startup founding credentials, and deep expertise in agentic RAG architectures that are central to AlpacaRelay's technical needs. Their work at Picard.law and Emission3 demonstrates exactly the 0-to-1 product ownership and full-stack AI engineering this role demands. The primary gaps are around a handful of specific tooling (LangSmith, LangFuse, MCP Servers, Kubernetes) and the absence of a public GitHub profile, which limits code quality verification. A technical interview is strongly recommended to validate engineering depth and close the tooling gap assessment — the overall profile suggests these gaps are likely closeable given their demonstrated adaptability across diverse frameworks.
Top Strengths
- ✓Genuine founding-engineer DNA with two active co-founder/founding-engineer roles shipped to real enterprise customers
- ✓Deep end-to-end ownership of complex AI pipelines — from OCR ingestion and fine-tuning to agentic orchestration and production deployment
- ✓Strong agentic AI stack breadth covering LlamaIndex, LangGraph, GraphRAG, RAG, vector search, and hybrid retrieval
- ✓Multimodal AI experience (text, vision, document AI, image generation) directly relevant to AlpacaRelay's content creation focus
- ✓Exceptional academic foundation (9.86 GPA, M.Sc. Industrial Mathematics) supporting strong ML and mathematical reasoning
Key Concerns
- !Missing explicit evidence of several key required tools (LangSmith, LangFuse, CrewAI, MCP Servers, Kubernetes) with no GitHub to cross-reference
- !LinkedIn date discrepancy (Emission3 showing 2026 start) and absence of education on LinkedIn warrant verification during screening
Culture Fit
Growth Potential
High
Salary Estimate
$80,000 - $110,000 USD (within posted range; India-based remote candidate may have different expectations — worth explicit discussion)
Assessment Reasoning
The candidate is marked FIT based on a score of 78. They meets approximately 70-75% of required skills with strong depth in the most critical areas — agentic AI, RAG architectures, LLM orchestration, Python, multimodal AI, and AWS — and demonstrates genuine founding-engineer experience that is rare and highly valuable for this role. The missing skills (LangSmith, LangFuse, CrewAI, MCP Servers, Kubernetes) are real gaps but represent tooling rather than foundational capability gaps; their broad framework experience strongly suggests rapid ramp-up. The absence of a GitHub profile and the LinkedIn date anomaly introduce moderate uncertainty, keeping confidence at 72 rather than higher. The recommendation is to advance to a technical interview with focused assessment on the missing tooling areas and a hands-on coding exercise to validate engineering craft before making a final hire decision.
Interview Focus Areas
Code Review
No direct code sample was provided, preventing a definitive assessment of code quality. However, the candidate's project descriptions — including a multi-stage open-source image AI pipeline, GraphRAG deployment at enterprise scale, and real-time IoT event processing — reflect architectural sophistication consistent with senior-level engineering. A technical screen with a code exercise is strongly recommended to fill this gap.
- +Drape open-source project demonstrates full-stack product thinking with a complex multi-stage AI pipeline shipped to production
- +Technical descriptions across resume show architectural maturity — hybrid retrieval, KV cache optimization, multi-tenant design — consistent with senior-level engineering
- -No code sample or GitHub profile provided, making direct code quality assessment impossible and limiting confidence in engineering craft evaluation
Experience Overview
8y total · 6y relevantThe candidate presents a compelling 8-year trajectory in production AI engineering with strong founding-engineer credentials at two startups. Their experience with agentic RAG, LlamaIndex, LangGraph, GraphRAG, multimodal AI, and AWS aligns well with the core role requirements. Some gaps exist in specific tooling (LangSmith, LangFuse, CrewAI, MCP, Kubernetes) but the breadth of their agentic AI work suggests these could be quickly bridged.
Matching Skills
Skills to Verify