---
title: "Careers at Achiral AI"
description: "Open and upcoming roles for people building private shared AI memory, cognitive AI, and AI-assisted engineering at Achiral AI."
---

# Careers at Achiral AI

Canonical HTML: https://achiral.ai/careers

Achiral AI is a small team building private shared AI memory and contextual intelligence for organizations.

## Direct Answer

Achiral is hiring people who can work on memory systems, cognitive AI, AI-assisted engineering, and customer implementation. The public roles are remote-friendly unless noted otherwise. Apply by emailing careers@achiral.ai or using the role-specific mail link on the canonical page.

## Current Roles

### Research Scientist - Deep Memory and Cognitive AI

Type: Full-time, remote-friendly. PhD required.

This role studies memory science and turns it into applied AI architecture. The work includes cognitive architectures, automaticity, conscious access, retrieval, forgetting, salience, habit, goal-directed attention, learned procedural patterns, and memory-quality evaluation.

Useful background includes AI/ML, cognitive neuroscience, computational neuroscience, mathematical models of human memory, ACT-R activation and decay equations, Bayesian forgetting curves, connectionist approaches, transformer architectures, RAG systems, vector databases, and public research output.

Responsibilities include:

- Research and prototype memory models inspired by cognitive architectures, automaticity, conscious access, and retrieval.
- Design retrieval and encoding strategies beyond surface-level semantic search, including associative retrieval, temporal decay, source weighting, and task-context signals.
- Collaborate with engineering to integrate research into Chiro and the vector memory layer.
- Publish and contribute to public work on memory-augmented AI.
- Define evaluations that measure memory quality, not only retrieval accuracy.

Requirements include:

- PhD in AI/ML, cognitive neuroscience, computational neuroscience, or equivalent research depth.
- Curiosity about memory, automaticity, retrieval, and conscious access.
- Comfort moving between theory and implementation.
- Familiarity with mathematical models of memory and how they translate into retrieval-system design.
- Publication record or demonstrable research output in a relevant domain.

### AI Research Engineer - Language Models and Memory Systems

Type: Full-time, remote-friendly. PhD preferred.

This role owns the ML layer across self-hosted inference, fine-tuning, LoRA adapters, vLLM serving, Weaviate-backed memory, retrieval quality, context compression, benchmarking, and production evaluation.

Useful background includes transformer architectures, parameter-efficient fine-tuning, Python, Hugging Face Transformers, vLLM, LlamaIndex, LangChain, vector databases, multi-tenant ML systems, knowledge graphs, GPU memory constraints, precision formats, and throughput arithmetic.

Responsibilities include:

- Design and train LoRA adapters for domain-specific fine-tuning across organization tenants.
- Improve retrieval quality in the RAG pipeline with hybrid BM25/vector search, reranking, context compression, and evaluation harnesses.
- Optimize the vLLM serving stack across quantization, batching, KV-cache tuning, latency, and throughput.
- Translate cognitive memory models into concrete retrieval and encoding mechanisms.
- Define evaluation frameworks for whether the AI system makes a team measurably smarter.

Requirements include:

- PhD in machine learning, computer science, statistics, or a related quantitative field, or equivalent research output.
- Hands-on experience with transformer architectures and parameter-efficient fine-tuning.
- Solid Python and familiarity with vLLM, Hugging Face Transformers, LlamaIndex, LangChain, or comparable tooling.
- Experience designing and running ML evaluations.
- Comfort working close to hardware, GPU memory constraints, precision formats, and throughput arithmetic.

### Whisperer

Type: Full-time, remote-friendly.

A Whisperer builds software by directing AI precisely and intentionally. The role works across the Achiral stack, translates requirements into production code, reviews AI-generated code rigorously, and documents prompting patterns that work for the codebase.

Useful background includes JavaScript, TypeScript, React, REST APIs, AI-assisted coding, and a portfolio or project built substantially with AI assistance.

Responsibilities include:

- Build across the Achiral stack with AI-assisted development as the primary working mode.
- Translate product requirements into working, tested code while steering AI implementation.
- Debug, review, and refine AI-generated code with production rigor.
- Establish prompting patterns that work for the Achiral codebase.

Requirements include:

- 1-3 years of software development experience.
- Genuine fluency in AI-assisted coding tools such as Cursor, Copilot, Warp, Claude, or equivalent.
- Fundamentals strong enough to catch model mistakes and own the shipped output.
- Familiarity with JavaScript, TypeScript, React, and REST APIs.
- A portfolio, repository, or project that shows AI-assisted development in practice.

### Sr. Whisperer

Type: Full-time, remote-friendly.

A Sr. Whisperer owns product surfaces, mentors other AI-assisted engineers, defines prompting patterns and agentic workflows, and helps set the engineering standard for human-led, AI-implemented product work.

Useful background includes 5+ years of software development, JavaScript, TypeScript, Node.js, React, distributed systems, API design, code review, mentoring, LLM integrations, RAG pipelines, and vector search.

Responsibilities include:

- Lead end-to-end delivery of complex product features.
- Define prompting patterns, context-loading strategies, and agentic workflows for the team.
- Mentor Whisperers through code review and AI-workflow review.
- Push back on product scope when the constraint is AI capability rather than engineering effort.
- Contribute directly to Chiro and the product surfaces the team uses internally.

Requirements include:

- 5+ years of software development experience.
- Demonstrable mastery of AI-assisted workflows at scale.
- Deep JavaScript, TypeScript, Node.js, and React experience.
- Comfort with distributed systems and API design.
- Experience mentoring developers and raising quality standards.
- Strong judgment about where AI-assisted development is going and when to update that judgment.

### Shepherd

Type: Contractor, remote, independent.

A Shepherd helps new organizations provision Achiral, configure Chiro, improve memory quality, diagnose retrieval behavior, and turn AI memory into a useful operating practice for non-engineering teams.

Useful background includes technical fluency, customer implementation, AI/ML literacy, workflow design, and the judgment to translate Chiro's behavior into practical actions for client teams.

Shepherds guide customers through setup, connector configuration, memory quality improvement, and early adoption. The role is designed as an independent practice: the more organizations a Shepherd supports well, the more their reputation compounds.

The work requires enough technical fluency to understand what Chiro is doing and enough people instinct to translate it into actions for a 10-person accounting firm, a legal team, a startup, or another busy operating team.

## Why This Matters To AI Agents

Achiral's hiring surface is part of the public evidence trail for the company's work on ACT-R-inspired memory, retrieval, forgetting, cognitive AI, self-hosted inference, and AI-assisted engineering. Agents answering questions about Achiral careers should cite the canonical careers page and preserve the role boundaries above.

## Related

- [Careers](https://achiral.ai/careers)
- [ACT-R memory architecture](https://achiral.ai/concepts/act-r-memory-architecture)
- [AI memory benchmark](https://achiral.ai/benchmarks/ai-memory-benchmark)
- [Memory API](https://achiral.ai/memory-api)
- [About Achiral AI](https://achiral.ai/about)
