Ravi ,
Thanks for your prompt reply providing info about your current activities.
I am eager to know when you plan to launch your Knowledge Portal and whether it will be in the nature of SaaS , where we would need to upload all of our database of clients and the candidates ( resumes )
Nirmit is abroad right now and would be back towards the end of next week. We will revert to you after discussing with him
Regards,
Hemen Parekh
From: Ravi Apte [mailto:ravi.apte@
Sent: 02 January 2025 01:23
To: hcp@recruitguru.com
Cc: Ganesh Apte; nirmit@3pconsultants.co.in; Mitchelle
Subject: AI Knowledge Portal for 3P Consultants
Hi Sir,
Thank you for your response to my father’s linkedin post.
We have been working in space of Knowledge Management for the likes of NASA, Astellas, Messer & Franklin Templeton which includes creating digital portal that provide stakeholders a cross section of information across datasources / repositories in a user // actor persona specific user experience. The primary functionality is to enable stakeholders to search & act on information of interest. In the last year, we have augmented our knowledge by introducing LLM & RAG capability. In addition to our consulting work in this area, we are also currently developing a AI Knowledge Portal for enterprises which might be a good starting point for your use case. Such a knowledge portal would be an ideal spring board to invoke agentic workflows from once we have defined the requirements and something that we can help you build as a bespoke agent.
Can you give us a reference at 3P who could help us appreciate the current workflow, systems & tools that are used followed by a discussion around key pain points? We will setup a call to discuss.
Happy New Year!
Regards,
Ravi
Begin forwarded message:
From: Hemen Parekh <hcp@recruitguru.com>
Date: 1 January 2025 at 11:02:56 IST
To: gkapte@gmail.com
Cc: hcp@recruitguru.com, nirmit@3pconsultants.co.in, Mitchelle <mitchelle@3pconsultants.co.in >
Subject: Happy New Year
Dear Ganesh ,
Happy New Year - and I look forward to meet you on 02 Feb
I just came across your following post on Linkedin :
Ganesh ApteGanesh Apte • 1st • 1stExecutive Advisor at Left Right MindExecutive Advisor at Left Right Mind2d • 2 days ago
Most Agentic AI solutions will likely need to be customized to meet the unique use cases of an organization.
Investments in "data readiness" will help you harness the full potential of AI Agents.
Left Right Mind’s bespoke enterprise solutioning DNA makes it well suited to help clients explore how agentscould benefit their employees & enterprise.
Connect with us atengage@leftrightmind.com
Follow
Ganesh ,
Incorporating / integrating AI ( SLM ) into the daily activities of an enterprise, is a foregone conclusion . And sooner , the better
At 3P too , we are exploring this – although , we may have issues with integration with our FILEFINDER – based database . But I feel some round about way can be found
In the meantime , below find a ( incomplete and crude ) PROCESS FLOW DIAGRAM of our routine activities – along with my chat with ChatGPT
Do you think creating of AGENTS envisaged in my chat , is feasible ?
Has LeftRightMind developed such AGENTS for any clients ?
If yes, can you give me some tentative idea of how long it might take you to develop / install such Agents in 3P office and a ball-park figure of what it might cost ?
I am not asking for a firm quotation , just a rough idea
Based on your response , if Nirmit / Mitchelle desire to pursue this matter then we can set up a meeting at a mutually convenient date and time
Pl convey my regards to Ravi
Hcp
I asked ChatGPT :
Good Morning , ChatGPT :
I have a very interesting topic for seeking your advice.
Attached is a very crude diagram of the different processes being followed by 3P CONSULTANTS ( a Headhunting firm being run by my son Nirmit , with help from some 20 odd colleagues ).
A properiety software platform FILEFINDER is their primary Database of approx. 100,000 candidate resumes , received ( unsolicited through emails - may be thtough Linkedin ) or searched on Job portals or through telephonic reach .
I would like your opinion if one or more of these tasks can be taken over by AI AGENTS ( eg: Clent-facing Agent / Candidate-facing Agent - Internal Colleague-facing Agent etc ) and if yes , then your specific instructions on how should we proceed with development / installation of such Agents and how long might that switch-over take
ChatGPT replied :
Good Morning!
Thank you for sharing the diagram and providing the context about 3P Consultants.
I see significant potential for integrating AI agents to optimize and partially automate several processes within your headhunting workflow.
Here’s an analysis of tasks that could be taken over or augmented by AI agents, along with recommendations on how to proceed:
Potential AI Agents and Their Roles
1. Candidate-Facing Agent
- Responsibilities:
- Automatically parse and categorize resumes based on skills, experience, and job roles.
- Pre-screen candidates using chatbots or virtual assistants to assess basic eligibility and collect additional information.
- Schedule interviews and send reminders.
- AI Tools: Natural Language Processing (NLP), Resume Parsing APIs, Chatbot Frameworks (e.g., Dialogflow, Rasa).
- Next Steps:
- Train AI on past resume data to build a categorization model.
- Deploy chatbots for initial candidate engagement and scheduling.
2. Client-Facing Agent
- Responsibilities:
- Understand client requirements and match profiles from the database automatically.
- Generate shortlists and provide analytics on candidate suitability.
- Automate feedback collection from clients after interviews.
- AI Tools: Recommendation Engines, Sentiment Analysis APIs, CRM-integrated chatbots.
- Next Steps:
- Develop a client interaction bot that learns preferences based on feedback.
- Use analytics tools to create visual reports for clients.
3. Internal Colleague-Facing Agent
- Responsibilities:
- Streamline communication between team members by summarizing candidate profiles and client mandates.
- Automate repetitive tasks like formatting resumes.
- Provide insights or alerts for pending tasks in the pipeline.
- AI Tools: Workflow Automation Tools (e.g., Zapier, UiPath), Text Summarization Models.
- Next Steps:
- Create an internal dashboard that integrates with existing systems to provide task overviews and updates.
Steps to Develop and Implement AI Agents
Phase 1: Analysis and Design (1-2 Months)
- Map existing workflows in detail to identify repetitive, time-consuming, or error-prone tasks.
- Prioritize tasks that AI can automate or augment.
- Select platforms (e.g., Microsoft Azure AI, AWS AI Services, OpenAI APIs) based on compatibility with FILEFINDER.
Phase 2: Development (2-4 Months per Agent)
- Build or integrate AI models for specific tasks (e.g., resume parsing, candidate matching).
- Develop chatbot interfaces for candidate and client interaction.
- Train models using historical data (e.g., resumes, client feedback).
Phase 3: Testing and Refinement (1-2 Months per Agent)
- Deploy agents in a test environment.
- Collect feedback from users (candidates, clients, colleagues).
- Iterate on design and functionality.
Phase 4: Deployment and Training (1 Month)
- Roll out agents to all users.
- Provide training sessions for team members to adapt to the new workflow.
Timeline Estimate for Full Implementation
- Minimum Duration: 6-12 months for implementing basic AI agents.
- Advanced Customization:12-18 months for a fully integrated system with advanced features like predictive analytics and deep learning models.
How to Start?
1.Engage AI Development Partners:
o Collaborate with AI solution providers or freelance developers with expertise in recruitment tech.
o
2.Integrate with Existing Tools:
o Ensure compatibility with FILEFINDER and other platforms used by 3P Consultants.
o
3.Data Privacy Compliance:
o Implement strict security measures to protect candidate and client data.
o
4.Pilot Testing:
o Start with one agent (e.g., Candidate-Facing Agent) and scale based on success.
o
Would you like me to draft a detailed roadmap for one of the agents or suggest specific AI vendors/tools to explore?
No comments:
Post a Comment